1. business and industrial
2. energy
3. oil

Krill
Biology, Ecology and Fisheries
Edited by
Inigo Everson
Leader, Marine Living Resources Section
British Antarctic Survey
Cambridge
b
Blackwell
Science
# 2000 Blackwell Science Ltd
Editorial Offices:
Osney Mead, Oxford OX2 0EL
25 John Street, London WC1N 2BS
23 Ainslie Place, Edinburgh EH3 6AJ
350 Main Street, Malden
MA 02148 5018, USA
54 University Street, Carlton
Victoria 3053, Australia
10, rue Casimir Delavigne
75006 Paris, France
Other Editorial Offices:
Blackwell Wissenschafts-Verlag GmbH
KurfuÈrstendamm 57
10707 Berlin, Germany
Blackwell Science KK
MG Kodenmacho Building
7±10 Kodenmacho Nihombashi
Chuo-ku, Tokyo 104, Japan
The right of the Author to be identified as the
Author of this Work has been asserted in
accordance with the Copyright, Designs and
Patents Act 1988.
All rights reserved. No part of this publication may
be reproduced, stored in a retrieval system, or
transmitted, in any form or by any means,
electronic, mechanical, photocopying, recording or
otherwise, except as permitted by the UK
Copyright, Designs and Patents Act 1988, without
the prior permission of the publisher.
First published 2000
Set in 10/13pt Times
by DP Photosetting, Aylesbury, Bucks
Printed and bound in Great Britain by
MPG Books Ltd, Bodmin, Cornwall
The Blackwell Science logo is a trade mark of
Blackwell Science Ltd, registered at the United
Kingdom Trade Marks Registry
DISTRIBUTORS
Marston Book Services Ltd
PO Box 269
Abingdon
Oxon OX14 4YN
(Orders: Tel: 01235 465500
Fax: 01235 465555)
USA
Blackwell Science, Inc.
Commerce Place
350 Main Street
Malden, MA 02148 5018
(Orders: Tel: 800 759 6102
781 388 8250
Fax: 781 388 8255)
Canada
Login Brothers Book Company
324 Saulteaux Crescent
Winnipeg, Manitoba R3J 3T2
(Orders: Tel: 204 837-2987
Fax: 204 837-3116)
Australia
Blackwell Science Pty Ltd
54 University Street
Carlton, Victoria 3053
(Orders: Tel: 03 9347 0300
Fax: 02 9347 5001)
A catalogue record for this title is available
from the British Library
ISBN 0-632-05565-0
Library of Congress
Cataloging-in-Publication Data
Krill: biology, ecology, and fisheries/edited by
Inigo Everson.
p. cm.Ð(Fish and aquatic resources series; 6)
Derived from a workshop held at the
Fisheries Center, University of British
Columbia, November 1994.
Includes bibliographical references (p. ).
ISBN 0-632-05565-0) (alk. paper)
1. Krill fisheriesÐCongresses. 2. KrillÐ
Congresses. I. Everson, Inigo. II. Series.
SH380.7.K75 2000
333.95'55913Ðdc21
00-045491
For further information on
Blackwell Science, visit our website:
www.blackwell-science.com
Contents
Series Foreword
Acknowledgements
List of Contributors
v
ix
x
Chapter 1
Introducing Krill
Inigo Everson
1
Chapter 2
Sampling Krill
2.1 Direct sampling
Jon Watkins
2.2 Acoustic estimation of krill abundance
Michael Macaulay
2.3 Biological observations
Inigo Everson
8
8
Chapter 3
20
33
Distribution and Standing Stock
3.1 Japanese waters
Yoshi Endo
3.2 Canadian waters
Inigo Everson
3.3 The Southern Ocean
Inigo Everson
40
40
Chapter 4
Aggregation and Vertical Migration
Jon Watkins
80
Chapter 5
Population Parameters
Volker Siegel and Stephen Nicol
103
Chapter 6
Reproduction in Euphausiacea
Robin Ross and Langdon Quetin
150
Chapter 7
Role of Krill in Marine Food Webs
7.1 Japanese waters
Yoshi Endo
182
182
53
63
iii
iv
Contents
7.2
7.3
Canadian waters
Ron Tanasichuk
The Southern Ocean
Inigo Everson
190
194
Chapter 8
Ecosystem Dynamics Involving Krill
Inigo Everson
202
Chapter 9
Krill Harvesting
Taro Ichii
228
Chapter 10
Products Derived from Krill
Stephen Nicol, Ian Forster and John Spence
262
Chapter 11
Management of Krill Fisheries in Japanese Waters
Yoshi Endo
284
Chapter 12
Management of Krill Fisheries in the Southern Ocean
Denzil Miller and David Agnew
300
Chapter 13
Management of Krill Fisheries in Canadian Waters
Inigo Everson
338
Chapter 14
Krill Fisheries and the Future
Inigo Everson, David Agnew and Denzil Miller
345
Glossary
349
Index
355
Taxonomic index
369
List of Abbreviations
371
Please note: the plate section falls between pp. 182 and 183.
Series Foreword
Fish researchers (a.k.a. fish freaks) like to explain, to the bemused bystander, how
fish have evolved an astonishing array of adaptations, so much so that it can be
difficult for them to comprehend why anyone would study anything else. Yet, at the
same time, fish are among the last wild creatures on our planet that are hunted by
humans for food. As a consequence, few today would fail to recognize that the
reconciliation of exploitation with the conservation of biodiversity provides a major
challenge to our current knowledge and expertise. Even evaluating the trade-offs
that are needed is a difficult task. Moreover, solving this pivotal issue calls for a
multidisciplinary conflation of fish physiology, biology and ecology with social
sciences such as economics and anthropology in order to probe new frontiers of
applied science. The Blackwell Science Series on Fish and Aquatic Resources is an
initiative aimed at providing key, peer-reviewed texts in this fast-moving field.
While bony fish stem from a great radiation that followed the invention of the
swimbladder in the Cretaceous period 100 million years ago, some fish groups, such
as the sharks, lungfish and sturgeons are more ancient beasts. Survivors from earlier
eras may be more widespread than we think: the deep sea coelacanths, formerly
known only from the Indian Ocean, have recently turned up in Indonesia. Also,
these fishes may be more effectively adapted to specialized niches than their ancient
body plans would suggest. For example, rays and angel sharks have perfected the art
of the ambush predator, while most cartilaginous fishes can detect electric discharges in the nerves of their prey.
Bony fish themselves have evolved into an amazing array of habitats and niches.
As well as the open sea, there are fish in lakes, ponds, rivers and rock pools; in
deserts, forests, mountains, the great deeps of the sea, and the extreme cold of the
Antarctic; in warm waters of high alkalinity or of low oxygen; and in habitats like
estuaries or mudflats, where their physiology is challenged by continuous change.
Air-breathing climbing perch (regularly found up trees), walking catfish and mangrove mudskippers are currently repeating the land invasion of their Carboniferous
ancestors. We can marvel at high-speed swimming adaptations in the fins, tails, gills
and muscles of marlins, sailfish and warm-blooded tunas; gliding and flapping flight
in several groups of fish; swinging, protrusible jaws providing suction-assisted
feeding that have evolved in parallel in groupers, carps and cods; parental care in
mouth-brooding cichlids; the birth of live young in many sharks, tooth carps,
rockfish and blennies; immense migrations in salmon, shads and tunas; and even the
so-called four-eyed fish, with eyes divided into upper air and lower water-adapted
sections.
v
vi
Series Foreword
In addition to food, recreation (and inspiration for us fish freaks), it has, moreover, recently been realized that fish are essential components of aquatic ecosystems that provide vital services to human communities. But, sadly, virtually all
sectors of the stunning biodiversity of fishes are at risk from human activities. In
fresh water, for example, the largest mass extinction event since the end of the
dinosaurs has occurred as the introduced Nile perch in Lake Victoria eliminated
over 100 species of endemic haplochromine fish. But, at the same time, precious
food and income from the Nile perch fishery was created in a miserably poor region.
In the oceans, we have barely begun to understand the profound changes that have
accompanied a vast expansion of human fishing over the past 100 years.
Krill: biology, ecology and fisheries edited by Inigo Everson from the British
Antarctic Survey team in Cambridge, UK. comprises the 6th book in the new
Blackwell Science Fish and Aquatic Resources Series. It includes 14 chapters
authored by permutations of 13 krill experts from 7 countries world-wide. The book,
like the workshop held at the Fisheries Centre, University of British Columbia, in
November 1995 from which it is derived (Pitcher & Chuenpagdee, 1995), aims to
address the problems and issues underlying the sustainable harvesting of krill. Five
inter-linked themes make up the focus of the book: krill ecology; krill sustainable
harvest; the ecological implications of krill harvesting; krill resource assessment
methods; and products and markets for krill.
The word krill comes from the Norwegian kril, meaning very small fish fry, but
has been used in English since 1907 for the shrimp-like food of baleen whales
(OED). Krill are defined here as the euphausids, a relatively small (about 85 species
world-wide) and uniform taxon of large pelagic shrimp-like filter-feeding crustaceans (adults are from 10±80 mm long). Most krill species inhabit the upper 400 m of
the ocean and shed their eggs into the sea, but a few (e.g. Nyctiphanes) have evolved
parental care of the eggs until hatching into nauplii. Many species exhibit diurnal
migration, approaching the surface to feed at night. Krill biomass world-wide is
thought to exceed 300 million tonnes and they are particularly abundant in North
Atlantic, Antarctic and North Pacific oceans. Krill, whose biomass often rivals that
of copepods in the plankton community, have a critical ecological role. They
comprise a critical link in oceanic food webs between their phytoplankton food and
their fish predators, many of which are commercially important fishes. It follows
that direct harvesting of such a pivotal component in the food web has an ecological
impact that must be evaluated if large-scale fisheries for krill are contemplated.
Most krill species are herbivorous, and in high latitudes krill consume many
diatoms, but many krill species can utilise zooplankton as well, and some (e.g.
Megayctiphanes) are carnivorous. So for most krill species, through the `cascade'
effect, phytoplankton might be expected to increase when krill are harvested by
humans. Organisms that feed on krill might decrease in abundance if krill are
harvested. These include many species of fish that are themselves the subjects of
substantial human fisheries. In the North Pacific, krill form an important component
of the diet of herring, salmon, pollack, sardine, mackerel and capelin, all of which
Series Foreword
vii
support important commercial fisheries. The role of krill as the food of squid is
thought to have been overestimated, but there is a fear that excessive harvest of krill
might lead to: algal blooms of under-harvested phytoplankton; reductions in krilldependent predators such as baleen whales and many fish; subsequent lower
abundance of commercial fish stocks through trophic cascade effects.
Abundance is the key input parameter for the evaluation of all harvest impacts
and for fishery regulation. The assessment and measurement of krill abundance
presents challenges to both existing sonar technology and to mathematical modelling. Estimating the biomass of krill in the plankton cannot be done with standard
fisheries acoustics technology, most of which has been designed for fish, which are
larger and have relatively high target strengths. Selection of equipment, selection of
frequency, target identification, calibration of gear and measurement error are
important topics in acoustic methodology for krill assessment.
Although krill growth appears to be well understood, modelling their population
dynamics is subject to considerable uncertainty, especially in recruitment and the
effect of social swarming behaviour. The design and analysis of surveys that can
provide robust estimates of krill abundance is therefore critical to success. Estimation of the potential yield of krill stocks leads to problems in demographics and in
the estimation of recruitment. Assessment and potential yield evaluations may
subsequently be used in developing suitable management measures for krill fisheries. Moreover, krill consumption by fish may be estimated and modeled, and the
impact of various levels of krill harvest forecast, estimates that must be incorporated
in any ecosystem-based assessment of krill fisheries.
The processing of organisms high in oils and pigments presents both technical
difficulties and economic opportunities to serve new and emerging markets.
Unconventional processing of krill products is developing unexpected new markets.
For example, krill pigments that may be used in the aquaculture feed industry fetch
much higher prices and profits than using krill for fishmeal or using krill tails for
human consumption, as has been done in Russia and Japan. Some krill enzymes
have a range of biomedical applications.
This book begins the vital process of evaluating the future and potential of krill
fisheries in the light of fisheries that are moving down the trophic levels of marine
food webs (Pauly et al. 1998 ) as traditional table fish resources become depleted
(Pitcher 2000). It will, I am sure, provide a definitive reference work on krill for
some time to come, and I am delighted to welcome it to the growing list of titles in
the Blackwell Science Fish and Aquatic Resources Series.
Professor Tony J. Pitcher
Editor, Blackwell Science Fish and Aquatic Resources Series
Director, Fisheries Centre, University of British Columbia, Vancouver, Canada
viii
Series Foreword
References
Pauly, D., Christensen, V., Dalsgaaard, J., Froese, R., and Torres, F. Jnr (1998) Fishing down
marine food webs. Science 279: 860±863.
Pitcher, T.J. (2000) Rebuilding as a New Goal for Fisheries Management: Reconstructing the
Past to Salvage The Future. Ecological Applications (in press).
Pitcher, T.J. and Chuenpagdee, R. (Eds) (1995) Harvesting Krill: Ecological impact,
assessment, products and markets. Fisheries Centre Research Reports 3(3): 82 pp.
Acknowledgement
The Series Editor, Editor and Authors are most grateful to the industrial companies
comprising the former Sustainable Ocean Resources Society of British Columbia
(now incorporated in the Institute for Pacific Ocean Science and Technology), the
British Columbia Ministry of Fisheries, and the Science Council of British Columbia
for sponsorship of the 1995 workshop from which this book on krill and its fisheries
grew.
Acknowledgements
Gestation for this book has been long and geographically widespread. It began with
a Workshop Meeting, organised by Tony Pitcher in November 1995 in Vancouver
when a number of the participants agreed to participate in the production of a book
on krill. Our timetable, agreed as being sensible at that time, went from being
realistic to elastic. The list of contributors changed although the majority stayed
with the project. To them all I extend my thanks for putting so much effort into the
venture and for providing me with encouragement to complete the project. I also
thank all those who provided review comments on the chapters. Various librarians
gave quick and courteous support and this is gratefully acknowledged. In particular
I thank Christine Phillips for her sisyphean attempts at teaching me the efficient use
of library systems and I am extremely grateful to her for her unerring skill at
locating documents from minimal or at times wrong information. Many other
people have helped with the project by freely discussing the subject matter at a wide
variety of meetings particularly within the CCAMLR arena. During my career two
people, Dick Laws and the late John Gulland, have provided considerable influence
and guidance over the years in matters both Antarctic and fisheries, this is gratefully
acknowledged. I also thank my wife, Diana, for allowing this book to permeate the
household for so long.
In addition I thank the following publishers for permission to reproduce figures
from their publications: Academic Press for permission to reproduce Fig. 7.3.3;
Alfred-Wegener for permission to reproduce Figs 8.1 and 8.2; Cambridge University Press for permission to reproduce Fig. 3.3.1; CCAMLR for permission to
reproduce Plate 2 and Figs 9.2.2 and 12.2.1; FAO for permission to reproduce Fig.
9.4.1; Fisheries Centre, UBC, Vancouver for permission to reproduce Fig. 3.2.3;
Inter-Research for permission to reproduce Figs 8.7b and 8.8; The Royal Society for
permission to reproduce Fig. 7.3.2; Scott Polar Research Institute for permission to
reproduce Figs 2.3.3. and 2.3.4 which originally appeared in a BIOMASS Handbook; Springer-Verlag for permission to reproduce Figs 2.3.2 which originally
appeared in Polar Biology. In addition I thank Hidehiro Kato and Steve Nicol for
permission to publisher their photographs as Plates 4 and 7 respectively.
Inigo Everson
ix
List of Contributors
David J. Agnew
Head, Falklands Group
Renewable Resources Assessment Group
T.H. Huxley School of Environment, Earth
Sciences and Engineering
Imperial College
Room 406
Royal School of Mines
Prince Consort Road
London SW7 2BP
Yoshi Endo
Laboratory of Biological Oceanography
Faculty of Agriculture
Tohoku University, Sendai
Japan
Inigo Everson
Leader, Marine Living Resources Section
British Antarctic Survey
Cambridge CB3 0ET
Ian Forster
The Oceanic Institute
41±202 Kalanianole Hwy
Waimanalo, H1 96795
USA
Taro Ichii
National Research Institute of Far Seas
Fisheries
Orido 5-7-1
Shimizu
Shizuoka 424
Japan
Michael C. Macaulay
Applied Physics Laboratory
College of Ocean and Fisheries Sciences
University of Washington
Seattle WA 98195
USA
Denzil G.M. Miller
Marine and Coastal Management
P/bag X2
Rogge Bay 8012
South Africa
Stephen Nicol
Australian Antarctic Division
Channel Highway
Kingston, Tasmania 7050
Australia
Langdon B. Quetin
Marine Science Institute
University of California
Santa Barbara Ca 93106
USA
Robin M. Ross
Marine Science Institute
University of California
Santa Barbara Ca 93106
USA
Volker Siegel
Institut fur Seefishcherei
Palmaille 9
22767 Hamburg
Germany
John Spence
Biozyme Systems, Inc
West Vancouver
BC V7V 1N6
Canada
R.W. Tanasichuk
Fisheries and Oceans Canada
Pacific Biological Station
Nanaimo
BC V9R 5K6
Canada
Jon L. Watkins
British Antarctic Survey
Cambridge CB3 0ET
x
Chapter 1
Introducing Krill
Inigo Everson
Krill, a term originally applied to `fish fry', is now taken to refer to euphausiids, a
group comprising over 80 species most of which are planktonic. They are widespread with examples to be found in all the oceans of the world. Their size and, in
places high numerical density, makes them of particular importance in some marine
ecosystems. That importance is enhanced in the case of the few species that are
commercially harvested.
Commercial harvesting and economic importance are the two criteria which,
when applied to euphausiids, provide the focus for this volume. This immediately
reduces the area of interest from all the oceans and seas of the world to the consideration of a few geographical regions. In the Pacific Ocean there is the area in the
north-west around Japan and in the north-east the coast of British Columbia. In the
North Atlantic, interest is concentrated on the Nova Scotian shelf and in the
Southern Ocean interest has been greatest in the Atlantic sector with lesser interest
in the Indian and Pacific Ocean sectors. Commercial interest in harvesting krill has
been reported from other regions such as the Mediterranean Sea and coastal Tasmania, although currently this appears to be of very limited extent. Several species
of krill aggregate into swarms and it is this swarming behaviour which makes them
attractive to commercial harvesting.
Focussing attention towards those species likely to be present in areas where
fishing is taking place invites a further restriction to the surface waters down to
approximately 500 m. Thus although there are different species present in deep
water they have little impact on or interaction with those that are found closer to the
surface. Applying these restrictions we are left with a small number of key species
and others whose ranges overlap with them geographically and vertically.
1.1
Euphausiid Identification
Accurate identification is central to many ecological investigations and is a topic
worthy of careful study. Convergent evolution in pelagic crustacea has resulted in a
variety of species that have similar superficial appearances but different phylogenies. Thus the first step must be to ensure that the specimen under consideration
is a Euphausiid and not, for example, a Decapod or Mysid. All Euphausiids have
1
2
Krill: Biology, Ecology and Fisheries
gills clearly visible below the carapace on thoracic segments seven and eight. The
morphology of a typical euphausiid is shown in Fig. 1.1. There are photophores or
luminous organs at the base of the abdominal pleopods, at the genital segment of the
cephalothorax and also near to the mouth parts. The photophores produce a blue
light which can often be seen in fresh caught live specimens while still in the net.
Fig. 1.1 Generalised view of a euphausiid to show the main morphological features mentioned in
the text and also that are relevant to identification.
There are good dichotomous keys to be found in Baker et al. (1990) and
Mauchline & Fisher (1969). Both of these are comprehensive and include
euphausiids world-wide, the former having a very good set of clear drawings. In
addition, for the north Atlantic and Southern Ocean Mauchline (1984, 1980a
respectively) provide good illustrated keys both of which are a truncated version of
the Mauchline & Fisher (1969) key. Although there is much commonality between
these keys it is worth consulting more than one, and also source literature, when
there is any doubt over the identity of a particular specimen. This is because there
are differences in the way individual features are illustrated and as an added bonus
many of the engraved plates of the early expedition reports, as well as possessing
considerable scientific merit, are also an art form frequently to be admired.
1.2
Broad-scale distribution
The following is a brief summary by way of general introduction to describe the
broad-scale distribution of euphausiids in the main areas of interest. More detailed
information on the key species of commercial interest and species that are likely to
be found associated with them will be found in Chapter 3. Further general information on distribution can be found in Mauchline & Fisher (1969) and Mauchline
Introducing Krill
3
(1980b); most of the following information has been summarised from these two
publications.
North Pacific Ocean
Several species are found on both eastern and western seaboards of the North
Pacific Ocean. The species of greatest commercial interest in the North Pacific
Ocean is undoubtedly Euphausia pacifica which is found across the Bering Sea,
through the Southern part of the Sea of Okhotsk and the Sea of Japan extending
southwards to about 308 N. The densest concentrations are generally found in the
North Pacific Drift and Aleutian Current extending southwards off the Californian
Coast and within the Californian Current. The southern limit for E. pacifica is at the
9.58C isotherm at a depth of 200 m. The only similar species with which it might be
confused is the much smaller E. nana whose distribution is restricted to Southern
Japan and the East China Sea.
Although rarely present in great numbers, Nematobrachion flexipes is found in
deep oceanic water of the North Pacific Ocean south of latitude 408 N. In addition,
between latitudes 358 and 558 N, Thysanopoda acutifrons is likely to be present
although its distribution appears to be limited by the 48C and 108C isotherm at
100 m depth. With a latitudinal span slightly further towards the equator, E. gibboides is found between 308 and 458 N across the Pacific Ocean and in the eastern
end of this range it extends down the Californian Current. Closer to the equator the
closely related E. hemigibba is probably restricted to deep oceanic waters between
latitudes 188 and 428 N and in the equatorial zone from 1608 E to 1108 W E. paragibba is likely to occur.
Tessarabrachion oculatum is a mesopelagic species found between latitudes 358
and 538 N. It is absent from the oceanic region south-east of Kamchatka and also
from the Bering Sea although it is found around the Aleutian Islands. The genus
Thysanoessa is characterised by species found in high latitudes. T. longipes is present in the North Pacific, Gulf of Alaska, Bering Sea, eastern region of the Beaufort
Sea and north of the Bering Strait. T. raschi, a neritic species, is found as far north as
808 N in the Arctic Ocean. In the Gulf of Alaska is T. spinifera another neritic
species whose distribution extends down to the Californian coast although it is not
found in the western Pacific. T. inermis has a latitudinal range extending from 638 N
in the Bering Sea south to latitude 43ë N and although it is present in the Arctic it is
not known whether there is continuity with Atlantic populations.
In the western Pacific region Pseudeuphausia sinica is neritic and confined to the
coastal regions of the East China and Southern Yellow Seas while Euphausia similis
extends northwards through the South and East China Seas and extends eastwards
of Japan in the Kuroshio Extension.
The final genus, Nematoscelis, is represented by three species. Nematobrachion
gracilis is found in the western Pacific in the Kuroshio Current and its extension
east of Japan. N. difficilis occurs across the Pacific Ocean between latitudes 358
4
Krill: Biology, Ecology and Fisheries
and 458 North and extends southwards to 208 N in the Californian Current. N.
microps is found north as far as 408 N in the western Pacific but is absent from
the eastern Pacific.
North Atlantic
The most important species is undoubtedly Meganyctiphanes norvegica which in the
west Atlantic is found northwards from Cape Hatteras at 358 North along the edge
of the continental slope to 708 North where it is found along the coasts of Labrador,
Baffin Island and West Greenland. It is also present around south and east
Greenland as far as 808 North. On the eastern side of the north Atlantic it extends
from the western end of the Barents Sea, and is found throughout the Norwegian
Sea, North Sea and Skagerrak. It is not however to be found in the Kattegat or
eastern English Channel. The main breeding areas are thought to be off the Gulf of
Maine, Gulf of St Lawrence, southwestern and southern Iceland and in the Norwegian Sea northwards to about 708 North.
Several other species are found on both sides of the Atlantic Ocean. On the
western Atlantic seaboard Thysanopoda acutifrons is found from the Gulf of
Maine to 708 North in the region of south-west Iceland and the Davis Strait while
on the eastern seaboard it is found from Gibraltar northwards to 708 North. In
the western Atlantic Thysanoessa longicaudata occurs from around the Gulf of
Maine at 358 North and extends as far as 708 North off West Greenland. On the
eastern side, although most common between latitudes 558 and 708 North, its
range overall extends from the Bay of Biscay to 838 North off north-east Greenland.
Nyctiphanes couchii is a neritic species which occurs in the eastern seaboard
above the slope and shelf of the north Atlantic between Gibraltar and 608 North.
Although present in the Irish Sea it does not appear to breed there or north of the
Skagerrak. It occurs in the Kattegat and is the only euphausiid to penetrate the
western Baltic although it does not breed there.
Thysanoessa raschi, another neritic species, is the commonest euphausiid off West
Greenland and extends northwards to about 708 and in extreme cases to 788 North.
It is present in the Gulf of St Lawrence and Gulf of Maine but is not found further
south than 408 North. On the eastern side of the Atlantic it is found around Iceland
and is common around Scotland as far as 558 North. It is also found along the
Norwegian coastline north to the Barents, White and Kara Seas. Thysanoessa
inermis occurs from the Gulf of Maine north as far as west Greenland at around 708
North. Off East Greenland it occurs as far north as 758 to 808 North although, except
in the Norwegian and Barents Seas, it does not breed north of 658 to 708 North. On
the eastern Atlantic seaboard it is not found south of about 508 North.
The only bottom dwelling euphausiid is Bentheuphausia amblyops which appears
to be widespread but is confined to water deeper than 1000 m from the equator to
latitude 468 15' North on both eastern and western sides of the Atlantic.
Introducing Krill
5
Southern Ocean
All the Southern Ocean euphausiids have a circumpolar distribution and are
broadly separated by their latitudinal ranges. The species commonly referred to as
the Antarctic Krill is Euphausia superba, a widespread species which frequently
swarms and is the subject of significant commercial fishing (Plate 1, facing p. 182).
Euphausia vallentini and E. longirostris are restricted to the subantarctic zone, the
former with a southern limit of the Antarctic Polar Front while the latter is found
only between latitudes 408 and 558 South. The southern limit of E. vallentini
approximates to the northern limit of E. triacantha a species having a circumpolar
distribution between latitudes 508 and 608 South. The southern limit of E. triacantha
overlaps the northern limit of E. superba and E. frigida both of which are only found
south of the Antarctic Polar Frontal Zone (APFZ). The most southerly species is E.
crystallorophias which is generally neritic and is restricted to the Antarctic continental shelf although some swarms have been found in the Scotia Sea (Brierley &
Brandon, 1999) and they are also present in the caldera of Deception Island.
Thysanoessa vicina and T. macrura are two species with a circumpolar distribution in the Southern Ocean south of about 508 to 558 South and extending into the
pack-ice zone. Although they are difficult to distinguish reliably without a binocular
microscope, these two species can be recognised from all others in the Southern
Ocean because their eyes are clearly in two parts with a constriction between. All
the other Southern Ocean euphausiids have circular eyes.
Other areas
There have been a few proposals to fish for krill in other areas. In a recent review
Nicol & Endo (1997, 1999) noted that there had been interest in fishing for krill in
the Mediterranean Sea, although it is not clear what the target species might be, and
also a proposal to fish for Nyctiphanes australis off Tasmania. In preparing this
chapter I was made aware of anecdotal information indicating that there was some
local fishing for Meganyctiphanes norvegica in Norwegian waters to provide feed for
the aquaculture industry; however, I have found nothing to substantiate these
suggestions.
1.3
Synopsis of the book
This book has been written primarily for postgraduate students, professional
scientists and administrators concerned with krill ecology and in particular fishery
management. Krill was initially seen as a ready source of protein to satisfy an
expanding world population although the expansion anticipated in the 1970s has
only partly materialised. Arising from this the main areas of commercial fishery
interest are relatively few as has already been mentioned.
6
Krill: Biology, Ecology and Fisheries
In Chapter 2 the methods of sampling krill are considered. Nets provide a method
of direct sampling from which information on the type and quality of the krill can be
determined. That method does have the disadvantage that it is time consuming.
Major improvement in the development of quantitative echo-sounders over the past
30 years has meant that this technique is the preferred method for covering large
areas and also for obtaining information on the local distribution of krill. Allied to
these sampling methods is the need to make biological observations on the krill and
these are discussed at the end of the chapter.
Having considered sampling methods the distribution and abundance of krill in
the main areas of commercial interest are discussed in Chapter 3. In the coastal
waters of Japan and Western Canada there are local fisheries while in the Southern
Ocean the fishery has a very much larger area over which it can be distributed.
Turning from the larger-scale distribution of krill considerations of swarming and
vertical migration are discussed in Chapter 4. It is these aggregation patterns which
make the krill suitable for commercial fishing and at the same time affect the
foraging behaviour of dependent species.
Keeping to a discussion of krill centred features in Chapter 5 we have a full
consideration of population parameters that can be used in developing population
models. The information in this chapter is wide ranging in order to provide the
reader with a view of the range of values likely to be encountered for each parameter. This broader view is retained as the reproduction of krill is considered in
Chapter 6. Here aspects of the development of germ cells are considered leading to
a discussion of the reproductive biology of various members of the group.
Having set the scene with regard to krill on their own, consideration is then given
in Chapter 7 to the role of krill in the ecosystems within which it is harvested. This
highlights the position of krill in the food chain. Against that background Chapter 8
considers ecosystem dynamics involving krill and considers the different time and
space scales that are of relevance in developing management advice. These range
from the broad scale of precautionary total allowable catches down to the interactions with the dependent species alluded to earlier.
The scene having been set, the emphasis changes wholly towards the krill fisheries
and their management. In Chapter 9 the krill fisheries are described in terms of
fishing methods, catch rates and history of the fisheries. A natural progression from
considerations of the capture of krill is discussion on the biochemical qualities that
can be utilised in the development of marketable products; these are discussed in
Chapter 10.
Moving from the fishery and its operation commercially Chapter 11 is the first to
consider fishery management, in this case in Japanese waters. The management
regime there takes account of market forces and environmental variation. Management of the krill fishery in the Southern Ocean is discussed in Chapter 12. The
Convention for the Conservation of Antarctic Marine Living Resources, being the
first international fishery agreement to require the management regime to take an
ecosystem approach, has required the introduction of novel approaches. These have
Introducing Krill
7
been considered by the authorities in determining how to manage the fisheries in
Canadian waters described in Chapter 13. The final chapter looks to the future by
considering what has been happening in the krill fisheries world-wide and considering how this might be brought forward as we move into the twenty-first century.
References
Baker, A. de C., Boden, B.P. & Brinton, E. (1990) A Practical Guide to the Euphausiids of the
World. British Museum (Natural History), London.
Brierley, A.S. & Brandon, M.A. (1999) Potential for long-distance dispersal of Euphausia
crystallorophias in fast current jets. Marine Biology 135, 77±82.
Mauchline, J.R. (1980a) Key for the identification of Antarctic euphausiids. BIOMASS
Handbook No 5, 4 pp.
Mauchline, J.R. (1980b) The biology of mysids and euphausiids. Advances in Marine Biology
18, 1±681.
Mauchline, J.R. (1984) Euphausiid, Stomatopod and Leptostracan Crustaceans. Keys and
Notes for the Identification of the Species. Synopses of the British Fauna. Edited by Doris
M. Kermack & R.S.K. Barnes, No 30. The Linnean Society of London and The Estuarine
and Brackish-Water Sciences Association.
Mauchline, J.R. & Fisher, L.R. (1969) The biology of euphausiids. Advances in Marine
Biology 7, 1±454.
Nicol, S. & Endo, Y. (1997) Krill fisheries of the world. FAO Fisheries Technical Paper 367,
100 pp.
Nicol, S. & Endo, Y. (1999) Krill fisheries: Development, management and ecosystem
implications. Aquatic Living Resources 12, 105±120.
Chapter 2
Sampling Krill
2.1
Direct sampling
Jon Watkins
There has been a continuing requirement to obtain samples of krill since the earliest
investigations into krill biology; in the case of Antarctic krill significant sampling
began in 1925 with the Discovery investigations (Kemp & Hardy, 1929). At that
time net sampling was the main way of obtaining information on distribution,
abundance, population demography and behaviour of euphausiids. Indeed, with the
exception of visual sightings of surface swarms of krill, it was not possible to know
where krill were without undertaking some form of net sampling. Since then the
techniques and equipment available to marine scientists have increased dramatically. Nowadays much of the information on distribution and abundance of krill can
be derived from remote sampling techniques, such as acoustics, as well as a variety
of direct sampling techniques.
The last 20 years have seen great advances with in situ observation and sampling
techniques which include diving (Hamner et al., 1983; O'Brien, 1987), remotely
operated vehicles (ROVs) (Marschall, 1988), underwater photography and video
(Guzman & Marin, 1982; O'Brien, 1987; Sameoto et al., 1993), and towed electronic
sensors such as the Optical Plankton Counter (OPC) (Herman et al., 1993). However, while these modern techniques provide data that could never be collected by
traditional net sampling, they have not eliminated the need for net sampling, rather
the emphasis for net sampling has changed.
Modern day net sampling programmes are now undertaken for the following
reasons:
. To validate or ground-truth remote sensing techniques.
. To collect animals for population demography and biological classification. For
example, to assess the length, maturity stage and age of the krill.
. To get live animals for laboratory studies of behaviour and physiology.
. To get estimates of biomass and distribution. This may be the only way for some
life stages such as the eggs and larvae to be sampled.
Such a wide range of objectives for net sampling means that many different types
8
Sampling Krill
9
of net have been developed and a range of strategies have been devised for using
these nets in the most efficient manner. In an ideal world a net would sample
whatever was in front of it and the resultant sample would be totally representative
of the local population. However, in reality all nets provide a biased sample. The
amount and type of bias will depend on the actual sampling gear and the chosen
sampling strategy in relation to the distribution, behaviour and life cycle stage of the
krill being studied.
No one net or sampling strategy is likely to be suitable for all purposes and so
arranging a modern-day net sampling programme can be an intimidating experience. This section discusses the major factors that need to be considered when
selecting an appropriate net and sampling strategy for a given project.
Problems with direct sampling
The ultimate aim of a sampling programme is to obtain a set of results that are
considered representative of the population under study. How we define representative and population are important. These will depend on each study and will
have a major effect on the degree of sampling bias that occurs or is deemed
acceptable (Watkins et al., 1990). Every type of sampler used will have a variety of
biases which ultimately affect the representativeness of the sampling. The level of
importance of these various biases will depend on the objectives of the sampling
programme. Sampling biases may be divided into three broad categories: availability of krill to be sampled, sampler avoidance and representativeness of sampling.
(a) Availability of krill to the sampler
Many species of krill undertake a diurnal vertical migration as discussed by Watkins
(Chapter 4). It is therefore quite possible that the proportion of a population that
could be found in front of the net at any one time will vary dramatically unless we
are able to sample the entire distribution range. The use of independent methods,
particularly acoustics, to determine the actual distribution of the krill has done much
to reduce the degree of error that may be caused by this general problem. Although
note that in the case of krill occurring near the surface many acoustic systems
regularly used do not sample this stratum (Everson & Bone, 1986a; Macaulay
Chapter 2).
(b) Net avoidance
It has long been known that not all the krill that are in front of a sampler will enter
the net (Tattersall, 1924; Mackintosh, 1934; Hardy, 1936). Initially avoidance was
surmised by comparing differences between day and night net catches. This was
thought to be a response primarily to visual cues and so the assessment of avoidance
was restricted to a relative comparison of catch rates (Brinton, 1967; Wiebe et al.,
10
Krill: Biology, Ecology and Fisheries
1982; Hovekamp, 1989). Most investigations of net avoidance have been field
studies; however, Fleminger & Clutter (1965) conducted experiments on a captive
population of mysids and copepods which provided substantially more control over
the results.
More recently evidence of significant net avoidance has come from comparisons
with other types of sampling, such as visual, photographic or acoustic estimates of
density. Thus Nicol (1986) compared the densities of Meganyctiphanes norvegica
surface swarms estimated from samples by three different methods:
. a bag-sampler which gave numbers up to 41 000 krill m±3;
. photographic methods, which gave numbers up to 770 000 krill m±3; and
. a plankton net which gave numbers that never exceeded 6 krill m±3.
Although precisely the same swarms were not sampled by each method it is clear
that the plankton net with estimated density orders of magnitude lower than the
other two methods must be providing a significant underestimate.
Acoustic density estimates have also been compared with net sample estimates
and the conclusion is that net samples are usually lower because of significant net
avoidance (Everson, 1987; Sameoto et al., 1993; Pauly et al., 1997; Watkins &
Murray, 1998). Of course with all such comparisons there is the possibility that the
independent sampling technique used may also be biased in some way. The relative
problems associated with visual and acoustic sampling have been discussed by
Watkins & Murray (1998). However, in the case of acoustic techniques there is an
additional advantage ± it is possible to see the krill actually avoiding the net. Thus
Everson & Bone (1986b) used an echo-sounder transducer mounted just in front of
the net mouth to observe how Antarctic krill moved out of the way to avoid a
rectangular midwater trawl (RMT8). Studies carried out by Sameoto et al. (1993) on
Meganyctiphanes norvegica used a variety of independent sampling techniques
together (ROV-mounted video, OPC, multi-frequency acoustics and net) to show
that net density estimates were 2±3 orders of magnitude less than those obtained
from the other sampling methods.
Net avoidance will affect the absolute number of krill entering the net. However, it may also affect the type of krill entering the net. Thus large krill, which
can swim faster than small krill, may be able to avoid the net more successfully
than small krill (Hovekamp, 1989). The type of net selectivity will depend on the
animals and the net used. For any net, selectivity is likely to vary for different
nets and this presents a major problem in comparing studies. It is interesting to
note that Bone (1986) found no obvious size difference in the krill caught by two
scientific nets, a rectangular midwater trawl (RMT8) and a large LonghurstHardy plankton recorder (LLHPR). Similarly Anon (1991) found no significant
differences in the catches of a number of scientific nets used in the BIOMASS
surveys of the South Atlantic in 1980. However, there were substantial differences between scientific nets and commercial trawls.
Sampling Krill
11
An interesting development to minimise the problems of net avoidance has been
put forward by Sameoto et al. (1993) who describe the use of strobe lighting on a
MOCNESS net. When a strobe light was shone ahead of the net during daytime
hauls the degree of net avoidance was significantly reduced. Net density estimates
with the light on were similar to those estimated with an OPC and much closer to the
acoustic estimates than when the light was not used.
(c) Degree of aggregation and the representativeness of sampling
All euphausiids exhibit some degree of heterogeneous distribution or patchiness.
The ultimate expression of this is the formation of discrete, high density swarms in
species such as Euphausia superba and Meganyctiphanes norvegica (see reviews by
Miller & Hampton, 1989; Siegel & Kalinowski, 1994). The formation of such
aggregations may have a number of effects on the sampling of krill. It has been
known for many years that plankton patchiness increases the sampling variability
(Silliman, 1946; Anraku, 1956; Wiebe & Holland, 1968), and it has generally been
determined that abundances have to differ by a factor of at least two to be considered significantly different. More recently it has been shown that the heterogeneous distribution of krill may affect more than just abundance estimates.
Watkins et al. (1986, 1990) showed that adjacent swarms frequently contain krill
which have different size or maturity status. Thus single net hauls are likely to
contain krill which are totally unrepresentative of the local population. For
example, it was calculated that to estimate the mean size of krill in an area with a
precision similar to that which would occur if there was no heterogeneity, would
require a minimum of 20 swarms to be sampled.
(d) The importance of sampling biases for different sampling objectives
In the previous paragraphs the various ways in which sampling biases may arise have
been discussed. However, the relative importance of these different biases will vary
according to the aims of the sampling programme. It is considered that net sampling
to estimate euphausiid biomass is subject to the greatest number of biases. In Table
2.1.1 the objectives of the sampling programme have been ranked against the
relevant biases. Thus availability of krill to the sampler is only a serious problem
when there is no independent method of detecting where the krill are in the water
column. Consequently if net sampling is used in conjunction with acoustic techniques then this is much less of a problem unless the krill are very close to the
surface and above the transducer, as might happen at night. When determining the
biological characteristics of the krill population it is less important whether some
krill avoid the net as long as there is no significant net selection. For acoustic
validation, bias caused by net selection is important if it is necessary to obtain a size
classification, but if the aim is only to confirm species identification then the
problem is reduced even more.
12
Krill: Biology, Ecology and Fisheries
Table 2.1.1 The importance of net sampling problems for different sampling programmes 5 5 ±
significant problem or bias, 5 ± minor problem or bias, 3 no problem.
Net sampling programme
Net sampling
problem
Availability of krill
Net avoidance
Selectivity
Degree of aggregation
Representatives of sampling
Estimates of
biomass
Biological
characteristics
Acoustic
validation
Live
animals
55
55
55
55
55
3
5
55
3
55
3
5
55
55
55
3
3
3
3
3
Sampler designs
Over the years the design of nets has changed in order to reduce the biases associated with net sampling and to cater for specific sampling objectives. General
principles for designing nets were thoroughly discussed by Clutter & Anraku (1968).
They concluded that samplers should have the following characteristics; they should
be:
.
.
.
.
.
as large as possible,
propelled as fast as possible at a constant speed,
capable of remaining at least 85% efficient throughout the haul,
free of forward obstructions forward of the mouth such as tow-lines, and
generally dark-coloured with no shiny metal components.
Some of these requirements are mutually exclusive, for instance net size and towing
speed tend to be inversely related.
Many modern nets have the towing cable attached to a point which is not directly
in front of the net (e.g. RMT ± Roe & Shale, 1979; MOCNESS ± Greene et al., 1998;
BIONESS ± Herman, 1988). The plummet net described by Hovekamp (1989) is
free of all obstructions forward of the mouth of the net as it has no bridles and
samples by free falling through the water column.
The size of many nets has increased greatly from that of the 1 m ring net used
during the Discovery expeditions and described by Kemp & Hardy (1929). Scientific
nets used to sample Antarctic krill often have a mouth area of 8 m2 (Roe & Shale,
1979). Other significant advances are the ability to take multiple samples within a
single haul, either using different sized nets (e.g. Bongo and RMT8+1 M) or using
multiple nets of the same size to sample different depth horizons (e.g. BIONESS,
MOCNESS, RMT). For sampling at a finer spatial resolution, the LLHPR is
capable of collecting up to 70 individual samples during a single haul (Bone, 1986).
Such modern multi-samplers are usually deployed with integrated oceanographic
instrument packages and real-time net to ship telemetry of key parameters such as
Sampling Krill
13
net depth, flowmeter readouts and net status (e.g. mouth angle, open/closed state).
With such equipment it is possible to target the net at particular targets detected on
a sonar or echo-sounder (Watkins & Brierley, in press). Most recently we see that
some net systems also have independent acoustic, video or particle-counter sensors
attached which provide important, additional information on the quality of the net
sampling (Sameoto et al., 1993).
Sampling strategies to obtain representative samples of population
This section on direct sampling techniques has demonstrated that net sampling
programmes are nowadays frequently used for purposes other than estimating
abundance or biomass of the population. This introduces a number of new priorities
when we design net sampling programmes. The plans for the CCAMLR synoptic
survey (SC-CAMLR, 1999) show some of the factors that need to be considered in a
modern net sampling programme which is carried out to support acoustic biomass
estimates and to describe the population demography. The various sampling
strategies that may be considered are discussed below.
(a) Daytime versus night-time net sampling
Evidence has already been presented on the large differences between day and
night-time net avoidance. Therefore a relatively simple strategy is to restrict net
sampling to the hours of darkness. Such a plan was used in the British Antarctic
Survey Core Programme, designed to monitor the interannual variability of krill at
South Georgia (Brierley et al., 1997). Two sampling criteria needed to be satisfied
for this programme. The first was to provide representative samples of krill for
length frequency and maturity stage analyses. The second was to determine the
species that were present in individual scattering layers. For this study station
sampling with nets was undertaken at night to minimise biases due to avoidance and
this was combined with acoustic surveys carried out during the day, when krill are
unlikely to be found close to the surface and consequently above a hull-mounted
echo-sounder transducer. The second question was addressed by undertaking a
series of daytime target validation hauls with the aim of providing a qualitative
indication of the dominant species that were present. A similar sampling strategy
was also adopted for the recent multi-national CCAMLR Synoptic Survey of the
Scotia Sea (SC-CAMLR, 1999).
(b) Station versus target fishing
Traditional net sampling programmes utilise a regular (Siegel, 1992) or random
sampling grid (Everson et al., 1996), however, in both cases the sampling is independent of the actual availability of krill. In such programmes the net(s) are usually
deployed over the entire depth range of occurrence of the krill. In the case of
14
Krill: Biology, Ecology and Fisheries
Euphausia superba, this is typically 0±250 m (Brierley et al., 1997). While such
sampling designs are important to enable a valid estimate of abundance they are less
than optimal if this is not the main aim of the sampling programme. If the aim is to
validate acoustic estimates or to assess the biological characteristics of swarms then
it is most efficient to target krill aggregations. Such target fishing (Ricketts et al.,
1992) is relatively easily done with modern equipment (echo-sounders, net-sondes,
and real-time data on net position and state).
We can see that depending on the aim it is advantageous to use either a target or
`random' net tow design. If the sampling programme has several aims we might
consider using both kinds of net tow. Unless it is possible to carry out a large number
of tows of each kind we must determine how representative our sampling may be.
For example recruitment indices for Euphausia superba have been calculated from
length density data collected with oblique net hauls carried out at regular stations
(de la Mare, 1994; Siegel et al., 1997). However, where the number of net hauls is
limited it may be necessary to combine target and station net haul data sets to
ensure that a sufficient number of net hauls are carried out to represent the
population adequately. Watkins (1999) describes the use of a composite index which
does this.
While it may be argued that targeted net hauls may give a biased estimate of the
population characteristics (Pauly et al., 1997; Watkins in press), a recent major
survey of Antarctic krill in the Indian Ocean sector of the Southern Ocean showed
that both types of net haul will tend to provide a representative description of the
population providing sufficient hauls are carried out (Fig. 2.1.1) (Pauly et al., 1997).
(c) How many swarms to sample?
I have already alluded to the problem of heterogeneity and sampling of krill. This
problem affects the sampling design for every sampling objective apart from collecting live krill for experiments (Table 2.1.1). Watkins et al. (1990) analysed surveys
over a number of different time- and space-scales and came up with some estimates
of the number of swarms that might need to be sampled. While such figures provide
a useful guide they demonstrate that it is very important that the trade off between
precision and number of samples should be assessed for any particular survey. One
implication that should be stressed for target hauls that are used to validate acoustic
surveys is that proximity of swarms does not imply similarity. Therefore while it
might appear more precise to link the results from single net hauls to small sections
of survey track adjacent to the net haul, the small-scale variability present means
that it is better to combine net hauls and generate an estimate of krill length distribution for a larger area. How is the size of such an area to be decided? One way
may be to use ordination techniques such as cluster analysis to define the areas on
the basis of a similarity matrix, such as has been done in survey analyses of population demographics of Euphausia superba (Siegel, 1988; Watkins et al. 1999).
There have been a number of papers that have considered how many krill should
% frequency
Sampling Krill
20
18
16
14
12
10
8
6
4
2
0
2
4
6
8
10
12
14
16
18
20
15
All transects
Target N = 7285
Routine N = 1564
Fig. 2.1.1 The length-frequency distribution of Euphausia superba caught in hauls directed at
acoustic targets (clear histogram) and in hauls taken at routine stations (shaded histogram). N
refers to total number of animals measured. Figure redrawn from Nicol et al. (in press) with permission from Elsevier Science.
be measured in any particular net. In situations where the sample is clearly unimodal with a narrow spread, a sample site of 50 may be adequate. However, if the
distribution is broad and polymodal then a sample size of 200 may be needed in
order to obtain a reasonable distribution.
A further consideration with regard to the sampling design of net haul surveys
concerns the trade-off between the number of hauls and duration of the individual
tows. Pennington & Volstad (1991, 1994) suggest that it is more efficient to reduce
the time of individual nets and carry out more net hauls than to carry out fewer,
longer net hauls. This is in line with the findings of Watkins et al. (1986) where it is
stressed that it is more important to sample more swarms of krill than to increase the
number of krill sampled from within the same swarm.
(d) Can we ever estimate absolute density from nets?
It has been shown that estimating density or abundance from nets is likely to be
more prone to bias than any other goal in a net sampling programme. The work of
Sameoto et al.(1993) shows that in the future it may be possible to reduce some of
the sampling biases caused by net avoidance. However, the large differences
16
Krill: Biology, Ecology and Fisheries
between the various methods used to estimate density using nets mean that where
possible other techniques should be used to estimate the density of krill. However, it
is only fair to point out that standardised net sampling programmes can have a very
important role in obtaining relative density estimates that can then be used to
monitor population changes (see for instance Siegel et al., 1997).
(e) Alternative sampling techniques to capture krill
Net sampling has been the traditional way to capture krill for measurement to
determine length frequency distributions for population demography. Depending
on the location and level of effort, net sampling programmes tend to be both
expensive and time consuming. It is therefore worth considering other ways in which
we may obtain samples of krill. One of the more interesting techniques is to use
predators to catch the krill and then sample the krill from the predators. Such
techniques have been widely used in the CCAMLR Ecosystem Monitoring Programme (CEMP) (SC-CAMLR, 1999). Predators used as samplers may be prone to
the same types of biases as nets. Of particular concern has been the selectivity of
predators for particular sizes or types of krill (see Nicol, 1993; Hill et al., 1996; Reid
et al., 1996). While this can undoubtably occur, Reid et al. (1999) showed that there
was good correspondence between the combined size distribution of Euphausia
superba taken by several different predator species (Macaroni penguins and Antarctic fur seals) and the size distribution derived from a series of net hauls conducted
in the same area and at the same time (Fig. 8.8).
One of the advantages of predator sampling is that it is possible to collect data
over longer temporal scales than may be possible in a ship-based sampling programme. Set against this is the problem of where predator samples may have come
from and the generally more restricted spatial coverage of the sampling. In the case
of sampling krill from predators at South Georgia, Reid et al. (1999) found a temporal change in size range of krill taken. While the temporal coverage of net-caught
krill was much less, the spatial coverage was significantly greater. It was possible to
see spatial trends in the data that would probably be reflected as temporal changes
when sampled from a fixed point past which a stream of krill was flowing (Watkins
1999).
References
Anon. (1991) Non-acoustic krill data analysis workshop. BIOMASS Rep. Ser. 66, 1±59.
Anraku, M. (1956) Some experiments on the variability of horizontal plankton hauls and on the
horizontal distribution of plankton in a limited area. Bull. Fac. Fish. Hokkaido Univ. 7, 1±16.
Bone, D.G. (1986) An LHPR system for adult Antarctic krill (Euphausia superba). Br.
Antarct. Surv. Bull. 73, 31±47.
Brierley, A.S., Watkins, J.L. & Murray, A.W.A. (1997) Interannual variability in krill
abundance at South Georgia. Mar. Ecol. Prog. Ser. 150, 87±98.
Sampling Krill
17
Brinton, E. (1967) Vertical migration and avoidance capability of euphausiids in the California Current. Limnol. Oceanogr. 12, 451±83
Clutter, R.I. & Anraku. M. (1968) Avoidance of samplers in zooplankton sampling. In
UNESCO Monographs on Oceanographic Methodology, 2: Zooplankton sampling (D.J.
Tranter, ed.), pp. 57±76. UNESCO Press, Paris.
de la Mare, W.K. (1994) Modelling krill recruitment. CCAMLR Sci. 1, 49±54.
Everson, I. (1987) Some aspects of the small scale distribution of Euphausia crystallorophias.
Polar Biology 8, 9±15.
Everson, I. & Bone, D.G. (1986a) Detection of krill (Euphausia superba) near the sea surface:
preliminary results using a towed upward-looking echo-sounder. British Antarctic Survey
Bulletin 72, 61±70.
Everson, I. & Bone, D.G. (1986b) Effectiveness of the RMT8 system for sampling krill
(Euphausia superba) swarms. Polar Biology 6, 83±90.
Everson, I., Bravington, M. & Goss, C. (1996) A combined acoustic and trawl survey for
efficiently estimating fish abundance. Fisheries Research 26, 75±91.
Fleminger, A. & Clutter, R.I. (1965) Avoidance of towed nets by zooplankton. Limnology
and Oceanography 10, 96±104.
Greene, C.H., Wiebe, P.H., Pershing, A.J. et al. (1998) Assessing the distribution and
abundance of zooplankton: a comparison of acoustic and net-sampling methods with DBAD MOCNESS. Deep-Sea Research II 45, 1219±37.
Guzman, O. & Marin, B. (1982) Hydroacoustic and photographic techniques applied to study
the behavior of krill. Mem. Natl Inst. Polar Res., Special Issue 27, 129±51.
Hamner, W.M., Hamner, P.P., Strand, S.W. & Gilmer, R.W. (1983) Behavior of Antarctic
krill, Euphausia superba: chemoreception, feeding, schooling, and molting. Science 220,
433±35.
Hardy, A.C. (1936) Observations on the uneven distribution of oceanic plankton. Discovery
Report 9, 511±38.
Herman, A.W. (1988) Simultaneous measurement of zooplankton and light attenuance with a
new optical plankton counter. Continental Shelf Research 8, 205±221.
Herman, A.W., Cochrane, N. A. & Sameoto, D.D. (1993) Detection and abundance estimation of euphausiids using an optical plankton counter. Marine Ecology Progress Series
94, 165±73.
Hill, H.J., Trathan, P.N., Croxall, J.P. & Watkins, J.L. (1996) A comparison of Antarctic krill
Euphausia superba caught by nets and taken by macaroni penguins Eudyptes chrysolophus:
evidence for selection? Marine Ecology Progress Series 140, 1±11.
Hovekamp, S. (1989) Avoidance of nets by Euphausia pacifica in Dabob Bay. Journal of
Plankton Research 11, 907±924.
Kemp, S. & Hardy, A.C. (1929) Discovery investigations, objects, equipment and methods.
Discovery Report 1, 141±232.
Mackintosh, N.A. (1934) Distribution of the macroplankton in the Atlantic sector of the
Antarctic. Discovery Rep. 9, 65±160.
Marschall, H.-P. (1988) The overwintering strategy of Antarctic krill under the pack-ice of the
Weddell Sea. Polar Biol. 9, 129±35.
Miller, D.G.M. & Hampton, I. (1989) Biomass Scientific Series, No. 9: Biology and ecology of
the Antarctic Krill (Euphausia superba Dena). A Review SCAR and SCJR, Scott Polar
Research Institute, Cambridge, England.
18
Krill: Biology, Ecology and Fisheries
Nicol, S. (1986) Shape, size and density of daytime surface swarms of the euphausiid Meganyctiphanes norvegica in the Bay of Fundy. J. Plankton Res. 8, 29±39.
Nicol, S. (1993) A comparison of Antarctic petrel (Thalassoica antarctica) diets with net
samples of Antarctic krill (Euphausia superba) taken from the Prydz Bay region. Polar
Biol. 13, 399±403.
Nicol, T., Kitchener, J., King, R., Hosie, G.W. & de la Mare, W.K. Population structure and
condition of Antarctic krill (Euphausia superba) off East Antarctica (80±1508E) during the
Austral summer of 1995/96. Deep-Sea Res. II, Topical Studies in Oceanography (in press).
O'Brien, D.P. (1987) Direct observations of the behaviour of Euphausia superba and
Euphauisa crystallorophias (Crustacea: Euphausiacea) under pack ice during the Antarctic
spring of 1985. J. Crustacean Biol. 7, 437±48.
Pauly, T., Nicol, S., de la Mare, W.K., Higginbottom, I. & Hosie, G. (1997). A comparison
between the estimated density of krill from an acoustic survey with that obtained by scientific nets on the same survey. CCAMLR, Hobart, WG-EMM-97/43.
Pennington, M. & Volstad, J.H. (1991) Optimum size of sampling unit for estimating the
density of marine populations. Biometrics 47, 717±23.
Pennington, M. & Volstad, J.H. (1994) Assessing the effect of intra-haul correlation and
variable density on estimates of population characteristics from marine surveys. Biometrics
50, 725±32.
Reid, K., Trathan, P.N., Croxall, J.P. & Hill, H.J. (1996) Krill caught by predators and nets:
differences between species and techniques. Marine Ecology Progress Series 140, 13±20.
Reid, K.G., Watkins, J.L., Croxall, J.P. & Murphy, E.J. (1999) Krill population dynamics at
South Georgia 1991-97, based on data from predators and nets. Marine Ecology Progress
Series 177, 103±114.
Ricketts, C., Watkins, J.L., Morris, D.J., Buchholz, F. & Priddle, J. (1992) An assessment of
the biological and acoustic characteristics of swarms of Antarctic krill. Deep-Sea Research
39, 359±71.
Roe, H.S.J. & Shale, D.M. (1979) A new multiple rectangular midwater trawl (RMT1+8 M)
and some modification to the Institute of Oceanographic Sciences (RMT1+8). Marine
Biology 50, 283±88.
Sameoto, D., Cochrane, N. & Herman, A. (1993) Convergence of acoustic, optical, and netcatch estimates of euphausiid abundance: use of artificial light to reduce net avoidance.
Canadian Journal of Fisheries and Aquatic Science 50, 334±46.
SC-CAMLR (1999). Report of the Working Group on ecosystem monitoring and management. In: Report of the eighteenth meeting of the Scientific Committee. CCAMLR, Hobart.
Siegel, V. (1988). A concept of seasonal variation of krill (Euphausia superba) distribution
and abundance west of the Antarctic peninsula. In Antarctic Ocean and Resources Variability (D. Sahrhage, ed.), pp. 219±30. Springer-Verlag, Berlin.
Siegel, V. (1992) Assessment of the krill (Euphausia superba) spawning stock off the Antarctic Peninsula. Arch. FischWiss. 41, 101±130.
Siegel, V. & Kalinowski, J. (1994). Krill demography and small-scale processes: a review. In
Southern Ocean Ecology: the BIOMASS perspective (S.Z. El-Sayed, ed.), pp. 145±63.
Cambridge University Press, Cambridge.
Siegel, V., de la Mare, W.K. & Loeb, V. (1997) Long-term monitoring of krill recruitment and
abundance indices in the Elephant Island area (Antarctic Peninsula). CCAMLR Sci. 4, 19±
35.
Sampling Krill
19
Silliman, R.P. (1946) A study of variability in plankton tow-net catches of Pacific Pilchard
(Sardinops caerulea) eggs. J. Mar. Res. 6, 74±83.
Tattersall, W.M. (1924) Crustacea, Pt. VIII. Euphausiacea. British Antarctic Terra Nova
Expedition 1910. Natural History Report Zoology 8, 1±36.
Watkins, J.L. (1999) A composite recruitment index to describe interannual changes in the
population structure of Antarctic krill at South Georgia. CCAMLR Sci. 6, 71±84.
Watkins, J.L. & Brierley, A.S. (in press) Verification of acoustic techniques used to identify
Antarctic krill. ICES J. mar. Sci.
Watkins, J.L. & Murray, A.W.A. (1998) Layers of Antarctic krill, Euphausia superba: are
they just long krill swarms? Mar. Biol. 131, 237±47.
Watkins, J.L., Morris, D.J., Ricketts, C. & Priddle, J. (1986) Differences between swarms of
Antarctic krill and some implications for sampling krill populations. Mar. Biol. 93, 137±46.
Watkins, J.L., Morris, D.J., Ricketts, C. & Murray, A.W.A. (1990) Sampling biological
characteristics of krill: effect of heterogeneous nature of swarms. Mar. Biol. 107, 409±415.
Watkins, J.L., Murray, A.W.A. & Daly, H.I. (1999) Variation in the distribution of Antarctic
krill Euphausia superba around South Georgia. Mar. Ecol. Prog. Ser 188, 149±60.
Wiebe, P.H. & Holland, R.H. (1968) Plankton patchiness: effects on repeated net tows.
Limnol. Oceanogr. 13, 315±21.
Wiebe, P.H., Boyd, S.H., Davis, B.M. & Cox, J.L. (1982) Avoidance of towed nets by the
euphausiid Nematoscelis megalops. Fish. Bull. 80, 75±91.
20
2.2
Krill: Biology, Ecology and Fisheries
Acoustic estimation of krill abundance
Michael Macaulay
Overview
Before an investigator employs acoustic methods to census krill, numerous factors
must be considered. Hydroacoustics, like any remote sensing method, requires a
good understanding of the sensing technology and its limitations. The investigator
will need to have confidence in what is being observed and how to best deploy
equipment for the task at hand. This section of the chapter will address some of the
ways krill behaviour alters their acoustic signature. The changeable acoustic characteristics of krill directly influence our choice of off-the-shelf hydroacoustic
instrumentation from among the different types available. Other items to be
covered include: what acoustic data can tell us about the distribution of krill in
relation to their environment, and how to manage the large quantities of data once
we have them. All these items have been identified (GLOBEC, 1991) as problems
that need consideration prior to using hydroacoustic methods to observe the highly
mobile krill. This section assumes that the reader is, at least passingly, familiar with
acoustic terminology. For those who want a more detailed study of acoustics, a good
reference on methods is given in MacLennan & Simmonds (1992).
Tools for acoustic observation
The acoustician can choose a single-beam system, dual-beam system, or a split-beam
system to conduct acoustic measurements. Single-beam systems were originally the
most commonly used, but split-beam systems have become the current standard for
echo integration estimation of biomass and other aspects of krill distribution. Dualbeam and split-beam systems are most frequently used to measure krill target
strength and to locate targets in an ensonified volume. To obtain the highest quality
volume scattering data and to make direct target strength measurements, it is now
common practice (Cochrane et al., 1991; Hewitt & Demer, 1993; Macaulay et al.,
1995) to deploy single-beam and dual-beam or split-beam systems operating at two
or more frequencies. Adjunct tools to improve localisation of concentrations of krill
are provided by searchlight sonar and side-scan sonar.
(a) Perfomance comparison
An investigator can measure target intensity (target strength) with single-, dual-, or
split-beam methods. Figure 2.2.1 shows a comparison of single-, dual-, and splitbeam systems and the possible appearance of echoes for the same set of targets for
each kind of system. The small X±Y plots below each example indicate the possible
intensity and/or phase signals for this set of targets with each type of system. In each
case, the first target is a small fish well off the main axis of all systems (target a in
Sampling Krill
Single-beam
Dual-beam
a
Split-beam
a
a
b
b
b
c
c
c
+
Angle
0
–
Range
Range
Wide
Left
0
–
Range
Phase 2
Range
0
Right
= phase centre
+
0
Aft
Phase 1
+
Angle
0
Narrow
Intensity
Intensity
Range
Intensity
Intensity
Fore
+
+
0
d
d
d
+
21
Range
Sum
Fig. 2.2.1 Comparison of the performance of single-beam, dual-beam, and split-beam systems on
a mixed population of targets. The shaded fish outlines in the single-beam illustration indicate
uncertainty whether target `a' is a small or large fish off axis or a small fish on axis, and whether
targets `c' and `d' are possibly one larger target. The dual-beam and split-beam systems properly
identify target `a' as a small fish off axis and targets `c' and `d' as two separate fish.
each column). In a single-beam system, it is not clear whether this target is a small
fish on axis or a larger fish off-axis (hence the shaded outlines in the single-beam
illustration). This positional and size uncertainty is a common problem with singlebeam systems. Dual-beam systems provide two channels of data, one from each
transducer (Fig. 2.2.1, second-column of the illustration). A target's position in the
acoustic beam is resolved using the signal intensity in relation to the beam patterns
for the narrow and wide portions of the transducer. The angular resolution of the
target's position forms an acceptance/rejection criterion, such that targets with
narrow beam intensity below a pre-established level (usually 3±6 dB down from the
wide beam intensity) are rejected as being too far off-axis to obtain a reliable
estimate of target strength. A dual-beam system resolves the first off-axis target,
22
Krill: Biology, Ecology and Fisheries
(that the single-beam system could not resolve) as probably a small target off-axis
(appears in wide-beam signal and poorly in the narrow-beam signal). The volume
scatter from a number of small targets at mid-range (target b in all columns) is
resolved as being off-axis in the dual-beam system. For the two targets at maximum
range (targets c and d in all columns), a dual-beam system may resolve whether this
is one larger target or two smaller targets. Split-beam systems (see the third column
of Fig. 2.2.1) resolve positional uncertainties by providing a more complete definition of the spatial distribution of targets (Traynor & Ehrenberg, 1979; MacLennan
& Simmonds, 1992). The split-beam system can resolve the angular bearing to each
of the example targets and would probably resolve the two targets at maximum
range as two separate targets, so long as the separation between the two is sufficient
to provide distinct phase angle differences. The phase angles for a target directly on
axis for the split-beam array would produce 08 phase angles for both axes, other
positions would produce phase angles with different magnitude and sign proportional to the target's position. Range-to-a-target is measured using the sum beam
signal. The two phase angles and the target range are then converted to a
mechanical offset angle in two axes and expressed in Cartesian or polar coordinates
to specify location of each identified target. The primary deciding factor between
using dual-beam or split-beam systems for measuring target strength is one of cost
(dual-beam systems are considerably less expensive). If precise positioning and wide
size-range of the target are important, then a split-beam system is always preferable.
Other types of acoustic systems, which have been or are being used for acoustic
observation, include: sector-scanning sonar, a means of electrically switching from
transducer element to transducer element that produces a precise localisation of
targets; searchlight sonar, in which a single transducer is mechanically scanned from
position to position to localise targets; and, side-scan sonar which uses a long,
narrow transducer with a very wide beam in one axis (often 908 or more at the
narrow dimension of the transducer), and a very narrow beam for the axis at right
angles to the first (often only 1±28 or less at the long dimension of the transducer).
These types of systems are frequently used to localise nets and schools, or patches of
fish or krill, relative to a fishing vessel for more effective catch strategies. In general,
these systems are not used for quantitative measurements because they are seldom
calibrated adequately to determine beam pattern, sensitivity, and source level
precisely. They are, however, very effective at improving the chances of being able
to locate and observe some key features of krill behaviour.
An additional type of acoustic system that aids in separating fish echoes from
other signals utilises a `chirp'. The production of a frequency band (the `chirp')
provides a way to separate individual targets from reverberation or background
noise. The system relies on signal processing methods to separate portions of the
returning echo into repetitive elements from those which are not (signal correlation), thereby improving recognition of discrete targets (i.e., short duration targets
like fish) from more uniform, slower-varying ones like krill patches and layers or
regions of acoustic noise.
Sampling Krill
23
Each of the types of systems described above are available as off-the-shelf systems
from a number of manufacturers. The investigator may also need to obtain training
on how to operate the associated data collection system. Having a clear sense of
precisely what needs to be measured and why ensures selecting the proper system or
systems for a particular investigation. A great deal of useful data can be collected with
simple, non-quantitative echo-sounders before investing in more quantitative ones.
The experience obtained may well save costly expenditure on the wrong equipment.
(b) System calibration
Acoustic systems are most often calibrated using standard targets (MacLennan, 1981;
Foote, 1982; Foote et al., 1987). Such targets are metal spheres of copper or tungsten
carbide whose size and composition determine their target strength. Other methods
of calibration have been used in the past (e.g. ANSI, 1972), but they are subject to
considerable error (Blue, 1984). Single frequencies are commonly employed, though
a more recent trend is to utilise more than one frequency simultaneously (Macaulay
et. al., 1995). Typically, frequencies from 38 kHz to 1±2 MHz have been applied in
hydroacoustic investigations with biological emphasis (Holliday & Pieper, 1980).
Survey sampling (where maximal depth range is desirable) requires the use of
frequencies below 200 kHz, while point or profile sampling (where ranges of tens of
metres or less is acceptable) can make use of frequencies above 200 kHz.
(c) Transducer design
Single-beam and dual-beam transducers are generally composed of single ceramic
elements or simple arrays of elements. In a split-beam system, a bi-axial transducer
is used to produce phase centres in a fore and aft plane (providing a fore and aft
axis), and left and right phase centres (providing a second axis at right angles to the
first one, see Fig. 2.2.1). The electrical phase difference for signals intersecting each
axis from a target is converted to a mechanical angle bearing relative to a central
axis perpendicular to the transducer face. These angular bearings, plus the slant
range to the target (provided by the sum beam), allow precise mapping of the
position of the target in Cartesian or polar coordinates, thereby removing the
positional ambiguity of the target inherent in dual-beam systems. Split-beam
transducers are constructed as an array of elements, especially at low (50 kHz or
less) frequencies. The transducer may also be constructed from a single piece of
ceramic divided into separate elements by slicing into several segments. There is no
clear preference between circular or rectangular designs so long as stable phase
centres can be established between the two halves of the transducer for each axis.
If the spacing and size of the transducer elements provide a distance between
phase centres of less than 1/2 wavelength for each axis, the transducer will produce
unambiguous phase angles over a wide range of directions away from the centre of
the transducer; however, this often results in a very wide beam pattern ± possibly too
24
Krill: Biology, Ecology and Fisheries
wide for a particular application. A more common practice in fisheries applications
is to locate the phase centres a few wavelengths apart by using a narrow-beam
transducer design. This minimises positional ambiguity by limiting acceptable phase
angles for valid targets to those targets known to be within the scope of the main
lobe of transducer sensitivity, and not detected by the side lobes. In all cases
(including single-beam and dual-beam), the transducer design needs to have minimal side lobes because the angular position and level of the side lobes establishes
the usable signal level and phase angle limits for determining target strength.
Asymmetrical transducer arrays (wider in one axis than the other) can provide
increased beam width for either axis to better fit the geometry of the environment.
Transducers with asymetrical beam patterns are often used in riverine or estuarine
environments where water depth may be a limiting factor.
(d) Near surface effects
The investigator should be aware of systematic limitations on the minimum range of
detection in utilising hydroacoustic methods for enumerating krill populations.
Because the acoustic properties of a signal near to a transducer are not well defined,
we must restrict the beginning of sampling to a range where they are well defined.
This range is approximately 10 transducer widths from the face of the transducer, or
on the order of 1 meter. In addition, the transducer is seldom located at the surface
and is usually several meters below, depending on whether it is hull-mounted or in a
towed body. The presence of noise generating bubbles, either from the ship or
weather-induced bubble entrainment, further limits the investigator's ability to
detect krill or other targets in the immediate surface layers of the water. In practical
terms, this means that the upper 10 m of the water column will be difficult or
impossible to examine quantitatively for populations of krill. This acoustic `blindspot' creates a systematic error in estimates of krill populations whenever significant
quantities of krill are located near surface. Fortunately, krill usually prefer deeper
depths, at least during daylight hours, although some individuals may be present
near-surface at all times. The diel migration of krill into the surface layers at night
may force acoustic surveys to be done during daylight hours only. In addition, krill
orientation and clumping in schools or patches is often very different between day
and night, and this can alter the strength of echo returned (Greenlaw et al., 1980;
Sameoto, 1980; Everson, 1982). These observations are likely the result of orientation on target strength (McGehee et al., 1998). Clearly, some knowledge of the
local behaviour of krill populations in any given study area is required before an
adequate survey design, using acoustic methods of sampling, can be developed.
Sound scatter versus reflection
Measurements of volume scattering distribution and intensity can provide estimates
of biomass for many small organisms that have low individual target strength. Low
Sampling Krill
25
target strength, ordinarily, would render the organism acoustically transparent.
Because many organisms frequently aggregate in large concentrations, the effect of
their combined scattering in an ensonified volume (i.e., volume scattering) produces
a measurable signal. Krill are a typical case in which volume scattering measurements provide an effective way to determine distribution and abundance. Figure
2.2.2 illustrates a typical situation where the only targets are fish, while Fig. 2.2.3
shows the same situation but with the presence of a krill layer. Size distribution of
krill can be extracted from multiple frequency single-beam systems (Pieper &
Holliday, 1984; Holliday & Pieper, 1989).
Fish only
Echo level (dB)
–80
–40
0
Fish targets
Bottom echo
Fig. 2.2.2 Illustration of how fish targets, diagrammatically shown in the left illustration, might
appear on an oscilloscope.
The estimation of krill abundance from volume scattering measurements (as is
done by echointegration) requires the consideration of factors seldom used in
acoustic estimation of fish abundance. Sound scatter from fish is predominantly in
the geometric region of sound reflection, that is, the target is large relative to the
wavelength of sound ensonifying it, and, there is little change in the strength of the
echo with frequency. Sound scatter from krill, on the other hand, is often in the
Rayleigh region such that the strength of the echo is strongly proportional to the size
of the target, or in a resonant region where echo strength varies dramatically with
changes in target size. Because krill are small and often nearly the same size as the
wavelength of sound ensonifying them, there is strong frequency dependence of the
level of scattering (see Fig. 2.2.4); this will create a similar frequency dependence in
estimates of abundance obtained using echo integration or echo counting. Concentrations of krill (in schools or patches) produce complex modes of sound scatter
26
Krill: Biology, Ecology and Fisheries
Fish and krill
Echo level (dB)
–80
–40
0
Krill layer
Krill layer
Fish targets
Bottom echo
Fig. 2.2.3 Illustration of how fish targets and krill, diagrammatically shown in the left illustration,
might appear on an oscilloscope. Note that the scope display shows the krill layer as a solid
envelope and fish targets as distinct signal spikes. Krill will seldom appear as individual targets,
rather they will be detected as a layer of volume scatter.
from the individuals comprising the layer (resonant effects). Krill, unlike fish,
exhibit a much stronger orientation dependence on the levels of sound scatter they
produce (MacLennan & Simmonds, 1992). The observer needs to be aware that
these effects combine to produce errors in the estimate of abundance obtained
acoustically. Recent papers by McGehee et al., 1998, and Traykovski et al., 1998,
using the distorted wave Born-Oppenheimer approximation model provide explicit
means of applying knowledge of the orientation of krill to adjusting target strength.
Unfortunately we are seldom in possession of adequate information to utilise such a
model. Currently, the linear model of Greene et al. (1991) and a multifrequency
method by Demer & Soule (1999) are routinely applied for rejecting multiple targets when making direct measurements of target strength in the field. Because krill
often actively avoid nets, estimates of abundance obtained strictly by net sampling
are subject to unknown or indeterminate biases; acoustic estimation of abundance
generally produces data with fewer biases and higher spatial resolution.
Target strength and identity
Target strength for zooplankton and other targets can be based on either measured
values (preferred), previously established values where direct measurement is not
possible, or on models like those developed by Stanton, (1989a, 1989b); Greene et
al., (1991); Macaulay, (1994); McGehee et al., (1998), or Stanton et al., (1998). A
researcher can establish preliminary target identity by observing echo signals on an
oscilloscope. Sound scatter produced by layers of zooplankton is often clearly
Sampling Krill
80 KhZ
27
43 MM Euphausiid
0
Directivity
–5
–10
–15
–20
–25
–90
–60
–30
0
30
60
90
Angle
0
40 KhZ
43 MM Euphausiid
Directivity
–5
–10
–15
–20
–25
–90
–60
–30
0
30
60
90
Angle
Fig. 2.2.4 Plot by frequency of normalised directivity (in dB) for a 43 mm euphausiid. Note the
strong central lobe of target strength (directivity) exhibited by the target at lower frequency and
the deep signal minima on either side of the peak. Individuals with a tilt angle of 308 upward
orientation or with 608 of downward orientation would be nearly one tenth as good a target as one
that was horizontal (i.e., 718±20 dB). Data from Kristensen & Dalen (1986) and solid curve from
Macaulay (1994).
distinct from individual fish echoes. Zooplankton have a broad, low-level signal
appearance as opposed to the spike-like appearance of fish targets (see Figs 2.2.3
and 2.2.4). Signal intensity is another descriptor, since fish schools, or layers, have a
much higher intensity than do zooplankton layers. Copepod layers have intensities
28
Krill: Biology, Ecology and Fisheries
of 745 dB to 743 dB at a maximum (equivalent density of individuals from net
samples was 40 000 individuals per m3; Macaulay et al., 1995), euphausiid layers are
frequently 10 dB higher (i.e., 735 dB, or 3000±5000 individuals per m3), while fish
schools are often 710 dB to 720 dB (i.e., 10±100 individuals per m3). Single echoes
from fish are on the order of 730 dB to 735 dB. A factor of 6 dB in intensity is
equivalent to a factor of two increase in signal strength; so, 10 dB is almost a factor
of four, and 20 dB is almost a factor of ten. See Miyashita et al., 1996 for an
example.)
The currently accepted target strength for Euphausia superba at 120 kHz is
TS(dB) = 7127.45 + 34.85 6 log10
(length in mm)
It is important that the investigator check such relationships carefully when applying
them to other species of euphausiid. The relationship of target strength to frequency
given in Greene et al. (1991) is useful as a check, bearing in mind that the inherent
linearity with frequency may not always apply (cf. Traykovski et al., 1998) at
frequencies higher than 200 kHz.
Comparisons of echo intensity can be used to separate likely targets, so long as the
types of targets are not excessively mixed. The use of multiple frequencies provides
significant assistance in this task (Pieper & Holliday, 1984; Saetersdal et al., 1984;
Macaulay et al., 1995). Miyashita et al. (1996) used comparisons of the scattering
from 38 kHz and 120 kHz to distinguish E. pacifica from fish. The selection of time
of day to do such comparative work (daylight hours, especially) permits separation
of targets due to their diel behaviour. At night, there is considerable mixing and this
necessitates using some bulk-average target strength to estimate abundance. An
investigator's sampling experience in a particular area provides information on what
ranges are reasonable for the resident target organisms. In general, the acoustic
characteristics (receiver sensitivity, beam pattern, type of system) will determine
what size of target, level of concentration, and range of observation is possible with
each type of acoustic system. For krill, this usually means narrow beam transducers
operating at 100±200 kHz and at ranges up to 250 m.
Krill behaviour
Krill often exhibit a strong tendency to aggregate in layers, schools, swarms, and
patches (Miller & Hampton, 1989). Consequently, special consideration must be
given to their behaviour in designing surveys to enumerate their populations.
Because krill are highly active swimmers, they are frequently capable of avoiding
many types of nets commonly used to collect biological samples. Increasing the
speed of the sampler or using a downward direction for sampling have proven to
increase the effectiveness of such samplers (Loeb et al., 1993). Ecological interactions between populations of krill (or other zooplankton) and the environment or
other organisms can be examined by spectral analysis of the spatial distributions
Sampling Krill
29
observed hydroacoustically (Macaulay et al., 1995). The spectral density of these
distributions often reveals scales of patchiness that are significantly different from
those for purely hydrographically distributed particles. Macaulay et al. (1995) suggests that there is a preferred patch size exhibited by many zooplankton and this
appears to be true for krill as well (Miller & Hampton, 1989).
The analysis of acoustic data from survey transects produces estimates of biomass
along the cruise track by distance. The distance covered in each integration interval
will depend on the speed of the vessel. In addition, vertical profiles of distribution
and abundance can be calculated for depth slices of selected thickness (usually 1±
10 m). These vertical profiles can then be used for statistical comparison with
hydrography and sampling by other investigations. A number of statistical methods
can be used to establish relationships of acoustic data (as an estimator of zooplankton abundance) to other biological sampling and to the prevailing hydrography. These methods include cluster analysis and stepwise discriminant function
analysis to develop indicators of biologically distinct areas (Macaulay et al., 1995),
and spectral analysis using a fast Fourier transform (FFT) to compare the spatial
scales of patch structure between areas (Platt & Denman, 1975; Weber et al., 1986;
Macaulay, 1991; Loeb et al., 1993). In addition to the collection of sound scattering
measurements of biogenic origin, researchers (Wishner et al., 1988; Macaulay, 1991;
Loeb et al., 1993; Macaulay et al., 1995; Beardsley et al., 1996) have shown that much
ecological information can be derived from the spatial and temporal patterns
observed in acoustically derived data.
Data management
To provide effective management information, survey data collection systems must
provide a number of items; among these are: real-time data display including
position; timely production of reports following field work; platform independence
of data formats and files; and manageable data volume. Flexible data export
capability allows convenient use of many commercial data analysis and display
packages. Because of the rapid evolution of the tools and platforms for analysis it
seems prudent not to be bound too tightly to any one system for data collection or
analysis (Macaulay, 1992).
The most important component of data management is the selection of a means to
store the information collected and to make it available to the user community.
While many database systems may have sufficient capacity and tools to satisfy data
storage and retrieval needs, there is an increasing tendency to use a geo-referenced
database (or Geographical Information System, GIS) to provide these functions.
There are many advantages to using regular databases and GIS systems together to
maximise the ability to extract, calculate, and display the survey data in the most
effective manner. There remain significant differences between display and information management tools available on different computer platforms, but generally,
once they are produced the resulting data products can be utilised across platforms.
30
Krill: Biology, Ecology and Fisheries
Conclusions
To make effective use of acoustics to measure the distribution and abundance of
krill an investigator should use more than one frequency and often more than one
type of system at the same time. Careful, regular calibration of the systems before
and after use is also required. In general, frequencies in the 100±200 kHz range are
the most useful for surveys, and these can be supplemented by other (higher frequency) systems where more detail is desired. Split-beam systems are preferred for
target-strength measurement work and can often be used simultaneously with single-beam systems. There is a suggestion from Traykovski et al. (1998) that there are
non-linearities in target-strength with frequency; care is needed when targetstrength is determined at one frequency and applied to a different frequency using
linear methods.
The most accurate results are obtained by target-strength measurements for each
frequency as well as net or other sampling for target identity and size distribution.
The researcher needs to examine the magnitude of near-surface concentrations
either by special sampling to determine the magnitude of such concentrations, or by
sampling at times when the krill are known to be less abundant. Krill behaviour will
often control survey design, and krill distribution can be used to reveal environmental structure (e.g. frontal zones due to currents or hydrography). Analysis of
distribution data from acoustic sampling can also be used to examine scales of
patchiness related to the environment that have great influence on predator distribution. This requires the use of spectral analysis and other statistical methods to
extract the complexities of the underlying distributions. Lastly, the investigator
needs to consider data management before conducting acoustic surveys. Acoustic
systems produce large quantities of detailed data on the abundance and distribution
of krill and other organisms. These data need to be reduced and stored in ways that
maximise their availability for the investigator to extract and analyse the results. To
this end, it is recommended that Geographical Information Systems (preferably
platform independent ones) be used to store the data. Almost all of the problems
that arise from the use of acoustic methods to estimate krill abundance are well
addressed by current software and hardware availability. After that, researchers
must ensure that careful consideration is given to understanding the effects of the
strengths and limitations that acoustic methods impose on the interpretation of
results.
References
American National Standards Institute Inc. (1972) Procedures for calibration of underwater
electroacoustic transducers. N.Y., Amer. Nat. Stds. Inst., Inc. N.Y. 10018.
Beardsley, R.C., Epstein, A.W., Chen, C., Wishner, K.F., Macaulay, M.C. & Kenney, R.D.
(1996) Spatial variability in zooplankton abundance near feeding Right Whales in the
Great South Channel. Deep-Sea Res. 45, 7±8.
Sampling Krill
31
Blue, J. E. (1984) Physical calibration. Rapp. P. -v. Reun. Cons. int. Explor. Mer. 184, 19±24.
Cochrane, N.A., Sameoto, D. Herman, A.W. & Neilson, J (1991) Multiple-frequency acoustic
backscattering and zooplankton aggregations in the inner Scotian shelf basins. Can. J. Fish.
Aquat. Sci. 48, 340±55.
Demer, D. & Soule, M. (1999) Improvements to the multiple-frequency method for in-situ
target strength measurements. CCAMLR WG-EMM 99/38.
Everson, I. (1982) Diurnal variations in mean volume backscattering strength of an Antarctic
krill (Euphausia superba) patch. J. Plank. Res. 4, 155±62.
Foote, K.G. (1982) Optimising copper spheres for precision calibration of hydroacoustic
equipment. J. Acous. Soc. Am. 71, 742±47.
Foote, K.G., Knudsen, H.P., Vestnes, G., MacLennan, D.N. & Simmonds, E.J. (1987) Calibration of Acoustic Instruments for Fish Density Estimation: a practical guide. ICES.
GLOBEC (1991) Workshop on acoustical technology and the integration of acoustical and
optical sampling methods. GLOBEC Report No. 4, p. 58. Joint Oceanogr. Inst. Inc.,
Washington DC.
Greene, C.H., Stanton, T.K., Wiebe, P.H. & McClatchie, S. (1991) Acoustic estimates of
Antarctic krill. Nature 349, 110.
Greenlaw, C.F., Johnson, R.K. & Pommeranz, T. (1980) Volume scattering strength predictions for Antarctic krill (Euphausia superba Dana). Meeresforsch 28, 48±55.
Hewitt, R.P. & Demer, D.A. (1993) Dispersion and abundance of Antarctic krill in the
vicinity of Elephant Island in the 1992 austral summer. Mar. Ecol. Prog. Ser. 99, 29±39.
Holliday, D.V. & Pieper, R.E. (1980) Volume scattering strengths and zooplankton distributions at acoustic frequencies between 0.5 and 3 mHz. J. Acoust. Soc. Am. 67, 135±46.
Holliday, D.V. & Pieper, R.E. (1989) Determination of zooplankton size and distribution
with multifrequency acoustic technology. J. Cons. int. Explor. Mer. 46, 52±61.
Loeb, V.J., Amos, A., Macaulay, M.C. & Wormuth, J.H. (1993) Krill stock distribution and
composition in the Elephant Island and King George Island areas, January±February, 1988.
Polar Biol. 13, 171±81.
Macaulay, M.C. (1991) AMLR PROGRAM: Spatial patterns in krill distribution and biomass near Elephant Island, Summer 1991. Antarctic J. 26, 205±206.
Macaulay, M.C. (1992) A hybrid database management system for the collection, display, and
statistical analysis of large and small scale hydrographic survey data. IEEE Proceedings,
OCEANS 92. October 1992. pp. 91±96.
Macaulay, M.C. (1994) A generalized target strength model for euphausiids, with applications
to other zooplankton. J. Acous. Soc. Amer. 95, 2452±66.
Macaulay, M.C., Wishner, K.F. & Daly, K.L. (1995) Acoustic scattering from zooplankton
and micronekton in relation to a whale feeding site near Georges bank, Cape Cod. Continental Shelf Res. 15, 509±537.
MacLennan, D.N. (1981) The theory of solid spheres as sonar calibration targets. Scottish
Fisheries Research Report, 22, 17.
MacLennan, D.N. & Simmonds, E.J. (1992) Fisheries acoustics. In Fish and Fisheries, Series
5, Chapman and Hall, London.
McGehee, D.E., O'Driscoll, R.L., Martin, V. & Traykovski, L. (1998) Effects of orientation
on acoustic scattering from Antarctic krill at 120 kHz. Deep-sea Res. II, 45, 1273±94.
Miller, D.G.M. & Hampton, I. (1989) Biology and ecology of the Antarctic krill. BIOMASS
Scientific Series.
32
Krill: Biology, Ecology and Fisheries
Miyashita, K., Aoki, I., Seno, K., Taki, K. & Ogishima, T. (1996) Acoustic identification of
isada krill, Euphausia pacifica Hansen, off the Sanriku coast, north-eastern Japan. Fish.
Oceanogr. 6, 266±71.
Pieper, R.E. & Holliday, D.V. (1984) Acoustic measurements of zooplankton distributions in
the sea. J. Cons. int. Explor. Mer. 41, 226±38.
Platt, T. & Denman, K.L. (1975) Spectral analysis in ecology. Annual Rev. Ecol. Syst. 61, 189±
210.
Saetersdal, G., Stromme, T. Bakken, B. & Piekutowski, L. (1984) Some observations on
frequency-dependent backscattering strength. FAO Fish. Rep. 300 (41), 150±56.
Sameoto, D.D. (1980) Quantitative measurements of euphausiids using a 120 kHz sounder
and their in situ orientation. Can. J. Fish. Aquat. Sci. 37, 693±702.
Stanton, T.K. (1989a) Sound scattering by cylinders of finite length. III. Deformed cylinders.
J. Acoust. Soc. Am. 86, 691±705.
Stanton, T.K. (1989b) Simple approximate formulas for backscattering of sound by spherical
and elongated objects. J. Acoust. Soc. Am. 86, 1499±1510.
Stanton, T.K., Chu, D. & Wiebe, P.H. (1998) Sound scattering by several zooplankton groups.
II: Scattering models. J. Acoust. Soc. Am. 103, 236±53.
Traykovski, L., Martin, V., O'Driscoll, R.L. & McGehee, D.E. (1998) Effect of orientation on
broadband acoustic scattering of Antarctic krill Euphausia superba: Implications for
inverting zooplankton spectral acoustic signatures for angle of orientation. J. Acoust. Soc.
Am. 104, 2121±35.
Traynor, J.J. & Ehrenberg, J.E. (1979) Evaluation of the dual-beam acoustic fish target
strength method. J. Fish. Res. Board Can. 36, 1065±71.
Weber, L.H., El-Sayed, S.Z. & Hampton, I. (1986) The variance spectra of phytoplankton,
krill, and water temperature in the Antarctic Ocean south of Africa. Deep-Sea Research 33,
1327±43.
Wishner, K., Durbin, E., Durbin, A., Macaulay, M., Winn, H. & Kenney, R. (1988) Copepod
patches and right whales in the Great South Channel off New England. Bull. Mar. Sci. 43,
825±44.
Sampling Krill
2.3
33
Biological observations
Inigo Everson
Behind any ecological study lies a great deal of basic observation on the species of
interest. For those species that are widespread in their distribution or that have been
studied by researchers from differing organisations it is quite common for different
criteria to be applied to the observations. Common cases where this has happened in
the past have been for length, maturity stage and feeding status. All of these are
important in population ecology and their value is enhanced when they can be easily
compared with results from other studies.
Length measurement
The morphology of euphausiids, with their long antennae and antennulae extending
forward, means that several definitions of length are possible. Even restricting the
definition to the anterior end of the main body poses difficulties because krill are
fragile and break easily, particularly in fixatives. Furthermore, they do not always lie
`straight' and although possessing a rigid exoskeleton, they can be compressed or
extended. Because it is so fundamental to many population studies length
measurement is worth some consideration here.
Mauchline (1980) when considering the Antarctic krill gave a reference
measurement of total body length which was defined as:
The lateral distance between the base of the eye-stalk and the posterior end of the
uropods, excluding their terminal setae.
In this context, the base of the eye-stalk is taken to mean the `anterior leading edge
of the carapace'. This is a measurement that is unfortunately quite tedious and
requires a low-power binocular microscope, consequently it is not surprising that it
has not been used for observations on fresh krill at sea.
A `straw poll' of colleagues known to the author and who have been involved in
measuring Antarctic krill at sea provided several approaches to what on the face of
it appears to be a relatively simple and standardised procedure. The standard
reference publication used by many workers is Mauchline (1980) who considered
the most practical total length measurement for E. superba, specified as Standard
Measure 1, to be:
The lateral or dorsal distance between the anterior tip of the rostrum and the
posterior end of the uropods, excluding their terminal setae.
The strict definition specified above is used by Australian workers (Nicol, pers.
comm.) although others take the posterior end as the telson tip rather than the
uropods. This is probably of minor importance in E. superba because when the
34
Krill: Biology, Ecology and Fisheries
uropods are laid lengthwise they extend to approximately the end of the telson. This
may not necessarily be true for all species.
The Standard 1 measurement, taking either the uropods or telson as the posterior
end, has been used by researchers from Australia, Japan and the United States
(personal communications from S. Nicol, T. Ichii and R. Ross). For the anterior end
such a definition is satisfactory for those species where the rostrum is small but if it is
of any significant size and likely to be prone to damage it may not be ideal. An
alternative that was used on the `Discovery Investigations' is:
from the front of the eye to the tip of the telson excluding the setae.
This has been used by researchers from Germany (V. Siegel, pers. comm.), South
Africa (D. Miller, pers comm) and the United Kingdom.
The question of length measurement was reviewed by CCAMLR for application
in its Observer Programme (CCAMLR, 1997). The requirement in this case was for
a method of measurement that could be undertaken at sea on commercial fishing
vessels. The definition given is from the front of the eye to the tip of the telson, the
`Discovery' standard. At the anterior end, the eye is more clearly visible than the
rostrum so that the two ends of the animal are thus clearly discernible. In making the
measurement it is probably best to hold the animal by the antennae and lay it on a
ruler or flat surface. Care needs to be taken due to the movement of the eyes on
their stalks as the animal is laid down. The different measurements are indicated in
Fig. 2.3.1.
Krill measurements have also been reported from samples obtained from predators. In such cases the individuals are often damaged and the options for measurement are restricted to the carapace, the largest hard part of the body. Mauchline
(1980) gives three measures which can be applied but favours the following definition of carapace length:
the dorsal distance between the anterior tip of the rostrum and mid-dorsal posterior edge of the carapace.
This measurement, indicated in Fig. 2.3.2. has been used by Australian and UK
workers (Nicol, pers. comm., Hill, 1990).
Even though the differences between results using the different methods are
small, of the order of one or two millimetres in the case of Antarctic krill, they do
introduce biases which could compromise the results based on comparisons between
locations or between years at the same locality. Precise standardisation would
therefore be an advantage for future studies.
A further point of similar concern is that of observer bias. Where groups of workers
are engaged in a series of measurements it is important to include an identifier for the
observer within the database and then to test for biases in the measurements before
drawing conclusions that rely on comparisons of size frequency distributions.
Sampling Krill
35
Fig. 2.3.1 Length measurements proposed for euphausiids. For explanation see text.
Fig. 2.3.2 Dorsal view of carapace to indicate carapace length measurement. (With permission
from and after Hill, 1990.)
Maturity stages
The secondary sexual organs of mature krill are reasonably easy to distinguish.
Mature male krill have petasmae present as modified endopods of the first pair of
pleopods (Fig. 2.3.3). They are swollen structures having a complex shape and are
used to transfer the spermatophores to the female. Paired vasa deferentia open
36
Krill: Biology, Ecology and Fisheries
Fig. 2.3.3 Mature male krill. Lateral view, upper diagram, to show location of petasma on
anteriormost pleopod and ventral view, lower diagram, to show location of ejaculatory duct at
posterior end of cephalothorax. With permission from SCAR. (After Makarov & Denys, 1981.)
externally on the ventral side of the thorax, their distal regions being swollen to form
ejaculatory ducts which are seen as paired red patches in the ventral integument.
In female krill the first pair of pleopods are not modified as sexual appendages.
On the ventral posterior region of the female cephalothorax there is a median trilobed structure, the thelycum. This is shown in Fig. 2.3.4. and is red in colour and
often has small white vesicles, spermatophores, attached to it. In unmated females
the red colour of the thelycum is often particularly bright.
The ability to determine whether krill are juvenile (in which case they would have
no external sexual characteristics visible) male or female is sufficient for many
Sampling Krill
37
Fig. 2.3.4 Ventral view of a mature female krill to show location of thelycum at posterior end of
cephalothorax. With permission from SCAR. (After Makarov & Denys, 1981.)
studies. When one is dealing with small species this is probably the limit which can
be achieved on fresh material at sea. For those species which grow to five or six
centimetres in length it is possible to categorise the krill into a series of maturity
stages with either the naked eye or else with the aid of a simple low-power magnifier. A scheme that has been used for many years for Antarctic krill was devised by
Makarov & Denys (1981) and this is set out below. This has been further developed
by the CCAMLR Scientific Observer Programme for observations on Antarctic
krill (CCAMLR, 1997). In common with the length measurements mentioned
above the system relies on freshly caught specimens and the manual provides the
following general guidelines. The basic Makarov & Denys (1981) descriptions are
given below.
Juveniles
Stage I
. Secondary sexual characteristics, petasma and thelycum are not visible.
Males
Stage II Sub-Adults
. Developing petasmae visible but not fully developed. Ejaculatory ducts are small
and not coloured.
Stage III Adults
. Petasmae fully developed; ejaculatory ducts red in colour and clearly visible
through the gills.
38
Krill: Biology, Ecology and Fisheries
. IIIA. Fully formed spermatophores are not present within the ejaculatory
ducts.
. IIIB. Fully formed spermatophores are present within the ejaculatory ducts.
Such spermatophores are easily ejected by exerting pressure on the ducts.
Females
Stage II Sub-Adults
. Developing thelycum is present but its colour is pale or colourless.
Stage III Adults
. The thelycum is fully developed, red in colour, and is clearly visible through the
gills.
. Stage IIIA. The thelycum bears no spermatophores and as a result, is very
brightly coloured. The body is not swollen.
. Stage IIIB. The thelycum is dirty red in colour owing to the presence of empty
spermatophores or, more rarely, has two opaque white points ± filled
spermatophores ± present. There is a free space between the ovary and the
walls of the body which is not swollen.
. Stage IIIC. Ovary fills whole thoracic space as well as the dorsal regions of the
first and second abdominal segments but the body is not noticeably swollen.
. Stage IIID. Thorax and first two abdominal segments are swollen owing to the
presence of the enlarged ovary which is clearly visible through the transparent integument.
. Stage IIIE. Thorax and first two abdominal segments are swollen but no
enlarged ovary is visible through the integument. Instead, the internal body
cavities are empty, the eggs having been laid. A small ovary is situated in the
middle of this space.
Systems of classifications such as this one where a continuous process is described
as a series of steps, always raises difficulties when `intermediate' forms are found.
However, since the key is being applied generally to determine the proportion of
krill of a given maturity stage that is present in the population then this problem is of
minor consequence. It is important, however, to operate a consistent policy with
regard to the `intermediates'. More detailed information on the gonad development
processes will be found in Ross & Quetin (Chapter 6).
Krill coloration
The feeding status of Antarctic krill, and other species that are filter feeding on
phytoplankton, can be investigated by examining the colour of the carapace and
hepatopancreas. As well as having an application in understanding ecosystem
interactions this analysis can be important in determining the quality of krill for
commercial processing, as discussed by Ichii, (Chapter 9) and Nicol et al., (Chapter
10).
Sampling Krill
39
The CCAMLR Scheme requires that the krill are compared against the coloured
standard shown in Plate 2 (facing p. 182). It is stressed that the krill should be
absolutely fresh and undamaged and that the analysis should take place in a cold
well-lit area. These criteria are important as autodigestion can be rapid and is
hastened in a warm environment. The krill themselves should be arranged into
colour groups and viewed against a white background. Four categories are described
for each of the carapace colour and the hepatopancreas colour and transparency.
References
CCAMLR (1997) Scientific Observers Manual. Commission for the Conservation of Antarctic Marine Living Resources, Hobart, Australia.
Hill, H. (1990) A new method for the measurement of Antarctic krill Euphausia superba
Dana from predator food samples. Polar Biol. 10, 317±20.
Makarov, R.R. & Denys, C. (1981) Stages of sexual maturity of Euphausia superba Dana.
BIOMASS Handbook No 11. 11 pp.
Mauchline, J.R. (1980) Measurement of body length of Euphausia superba Dana. BIOMASS
Handbook No 4. 9 pp.
Chapter 3
Distribution and Standing Stock
3.1
Japanese waters
Yoshi Endo
Euphausia pacifica is widely distributed in the North Pacific Ocean. In the Japanese
waters of the Pacific Ocean, it occurs as far south as Suruga Bay, 34850'N (Sawamoto, 1992) and extends northwards as far as the southwestern area of the Okhotsk
Sea (Ponomareva, 1963; Ohtsuki, 1975). It is present over almost the entire Sea of
Japan extending northwards as far as the southern part of the Gulf of Tartary
(Komaki & Matsue, 1958; Ponomareva, 1963).
Oceanographic conditions and the distribution of Euphausia pacifica
A part of the Oyashio Current flows southward along the east coast of Honshu
Island as the first and second Oyashio branches. Another part returns northeastward as the Oyashio Extension, forming the Oyashio Front with the high salinity
waters originating mostly from the subtropical gyre (Fig. 3.1.1) (Yasuda et al., 1996).
The first and second Oyashio branches, modified through contact with Tsugaru
Warm Current water and warm-core rings cut off from the Kuroshio Extension,
meet with the Kuroshio Extension and form the Kuroshio Front. The region
between the Oyashio and Kuroshio waters is referred to as the mixed water region,
perturbed area or transition zone (Yasuda et al., 1996).
The relationship between oceanographic conditions and fishable E. pacifica
aggregations in Sanriku waters is well documented by Odate (1991) and Kodama &
Izumi (1994). Using information on fishing conditions, Kodama & Izumi (1994)
reported that the fishing grounds are formed when the water mass whose temperature is in the range 78 to 88C is pushed to the shore by the approach of the first
Oyashio branch, which has a temperature of < 58C from the surface to the bottom.
The authors speculated that the optimal temperature for E. pacifica in the fishing
season is 78 to 88C and that this species avoids water colder than 58C or warmer than
108C. Odate (1991) also concluded that the main distributional area of the species in
the Sanriku and Joban regions is confined to waters where the temperature is
between 58 and 108C. The lower temperature limit for the species, based on the
fishing condition data in the area, is around 48C.
40
Distribution and Standing Stock
41
50
Western Subarctic
Gyre
I sl
an
ds
Ventilation
Okhotsk Sea
ril
45
SC
Ku
r o nt
rctic F
Suba
Transition Domain
Hokkaido
40
Japan Sea
Subarctic Boundary
W
O2
a
shim
Tsu rent
r
u
C
35
O1 WCR
Kuroshio Bifurcation
Front
WCR
KF
Honshu
Kuroshio Extension
Subtropical Mode Water
CCR
CCR
30
25
130
135
140
145
150
155
160
165
170
Fig. 3.1.1 Schematic representation of the current and frontal systems in Japanese waters
(Yasuda et al., 1996). TW: Tsugaru Warm Current; O1: Oyashio first branch (Coastal Oyashio
Intrusion); O2: Oyashio second branch (Offshore Oyashio Intrusion); WCR: warm-core ring, CCR:
cold-core ring; KF: Kuroshio Front.
More recently, the distribution of the species has been investigated throughout
the year in the Sanriku and Joban areas (Taki et al., 1996; Taki & Ogishima, 1997).
Based on samples collected with a beam trawl and an Ocean Research Institute net
(Omori, 1965), with a mouth diameter of 1.6 m and a mesh size of 330 mm, both of
which were fished along a transect line off Onagawa by day and night every two
months from April 1993 to February 1994, Taki et al. (1996) reported seasonal
change in horizontal distribution of E. pacifica. In April, the krill were distributed
pelagically mainly in coastal areas from 100 and 150 m depth. In other seasons the
main distributional area shifted offshore within the depth range from 200 to 300 m.
During the period from August to December, adults were mostly benthopelagic
during daytime, but immature animals had a more pelagic distribution. Furcilia
larvae, which occurred throughout the year, but were most abundant in April, were
entirely pelagic. The bottom temperature was 28 to 88C when benthopelagic krill
appeared. Benthopelagic concentrations rarely occurred when the bottom temperature exceeded 108C.
Taki & Ogishima (1997) reported the seasonal distribution of all developmental
stages as well as growth of E. pacifica using data from monthly samples, over the
period from February to November 1994. These were obtained using Norpac nets
42
Krill: Biology, Ecology and Fisheries
(Motoda, 1957), with a mouth diameter of 45 cm and a mesh size of 330 mm, hauled
vertically from 150 m to the surface from wider areas of Sanriku and Joban waters.
Although this was the most extensive krill survey conducted in the area, the net was
not large enough to collect adults effectively and the sampling depth did not cover
the whole daytime distributional range of the species. Taki & Ogishima (1997)
found that all the developmental stages appeared throughout the survey period and
were most abundant in the transition zone where the water temperature was in the
range from 58 to l08C at 100 m depth.
In February, eggs and early larvae did not occur in the area under the influence of
the Oyashio Current where the water temperature is < 58C at a depth of 100 m.
However, they were present in water of 58 to 108C at that depth. They occurred in
colder waters later in the season at which time they were concentrated in water of
about 58C at 100 m depth. Most spawning activity appears to have occurred in
April±June, as has been suggested by previous workers (see Nicol & Endo, 1997).
In 1996, the North Pacific Krill Resources Research Group undertook a cooperative winter±spring survey of the horizontal distribution of euphausiids in
Sanriku waters. The improved larva net with a mouth diameter of 1.3 m and a mesh
size of 0.45 mm (Watanabe, 1988) was hauled obliquely from 150 m to the surface at
night. During January and February, the abundance of adult E. pacifica was higher
in Oyashio than in the waters related to Kuroshio (including the Tsugaru Warm
Current water and warm water related to the Kuroshio Extension). It increased by
an order of magnitude in the Kuroshio area in late March to late April. This suggests
that there was a shift in horizontal distribution of the species from a cold to a
warmer regime just before the onset of the fishing season (Taki et al. in preparation).
E. pacifica is known to form dense swarms during the fishing season (Komaki,
1967; Terazaki, 1980; Endo, 1984; Hanamura et al., 1984). Juveniles of the species
also swarm in the Kuroshio Front at night (Nishikawa et al., 1995). Nishikawa et al.
(1995) also investigated the distribution of euphausiids in the Kuroshio Front and
warm water tongue in the Kashima-nada Sea, the southernmost distributional area
of E. pacifica (Fig. 3.1.2). This warm tongue of Kuroshio water separated from the
main current and stretched to the north along the continental shelf in the Kashimanada Sea. In the Kuroshio area, E. similis and Nematoscelis microps were dominant.
E. pacifica was the dominant species in the front area and the warm water tongue,
comprising more than 95% of the total number of euphausiids in the frontal area.
They were present in numerical densities up to 238 individuals m±3 in the frontal
area. Surface aggregations of juvenile E. pacifica, of around 8 mm body length, have
also been observed with the naked eye at night in the frontal area. Nishikawa et al.
(1995) concluded that the convergent flow pattern at the front may be an important
factor leading to the formation of surface aggregations of juvenile E. pacifica.
Taki et al. (1996) reported that both juvenile and adult E. pacifica were abundant
in the water column when the average temperature was 78 to 148C (Fig. 3.1.3). If the
surface water temperature does not exceed 208C, they migrate to the surface layer at
Distribution and Standing Stock
43
Fig. 3.1.2 Locations of sampling stations and surface water temperature gradients in the Kashimanada Sea (Nishikawa et al., 1995). A warm water tongue is delineated by 228C contour line. O:
Oyashio Current; S: Sanriku coast; T: Transition zone; K: Kuroshio Current.
night (Iguchi et al., 1993). The occurrence of juvenile E. pacifica in the Kuroshio
front and the warm water tongue is not surprising in view of the water temperature.
Warm core rings and Euphausia pacifica distribution
Warm-core rings shed from the Kuroshio Extension are reported to have profound
effects on the distribution of copepods (Hattori, 1991, 1993) and chaetognaths
(Terazaki, 1992) and the same has proved to be true for the euphausiid population.
Taki (1998a) investigated the euphausiid distribution in and around warm-core ring
93A which is believed to have been formed in 1993. He used MOCNESS-1 (Wiebe
et al., 1985) to collect euphausiids from 5 depth layers down to 1000 m at 6 stations
along 39830'N. Five warm-water euphausiid species, E. gibboides, E. similis, E.
hemigibba, Nematobrachion flexipes and Nematoscelis gracilis, were present in
warm-core ring 93A; the temperature of the mixed layer formed by the winter
convection was 128 to 138C in July 1995. At that time E. pacifica were scarce,
representing only 0 to 8% numerically of the total adult euphausiids in the centre of
44
Krill: Biology, Ecology and Fisheries
Fig. 3.1.3 Occurrence of furcilia, immature and adult of Euphausia pacifica during the day (left
panels) and night (right panels) at stations off Onagawa plotted on T±S coordinates (Taki et al.,
1996).
the ring, but they were abundant outside and beneath the ring accounting for up to
90% of juvenile and adult euphausiids. In the next year, E. pacifica were still scarce
in 93A, but abundant outside of the ring, although in January 1996 the mixed-layer
temperature had decreased to 108 to 118C (Taki et al. in preparation). In an old
warm-core ring with an observed temperature of 48 to 58C, Endo (1993) reported
that E. pacifica dominated the euphausiid population accounting for more than 95%
numerically of the catches; these included very small numbers of E. gibboides, the
sole warm-water species. Further studies are needed to determine the processes
associated with regulating the incursion of E. pacifica into warm-core rings.
Benthopelagic aggregation
At the end of the fishing season, surface aggregations of E. pacifica, which earlier in
the season are present in the coastal waters of Ibaraki Prefecture, disappear as the
surface water temperature increases. At that time, aggregations were distributed in
the cooler mid-water zone (Suzuki, 1986). Since this discovery, mid-water swarms as
Distribution and Standing Stock
45
well as benthopelagic and surface swarms have been targeted by the fishery
(Nakamura, 1992). According to Nakamura (1992), surface aggregations were
formed in years when cold water (58 to 78C) prevailed from the surface to the
bottom, whereas benthopelagic aggregations were formed when the water column
was stratified with warm water > l08C at the surface overlying a thick cold (68 to 98C)
deeper layer. When the water column was stratified and the cold water layer was
thin, neither surface aggregations nor benthopelagic aggregations appeared.
Monthly investigations of the benthopelagic E. pacifica have been made since
September 1989 at five stations off Nakaminato (36820'N) (Nakamura, 1991; Ebisawa, 1994). A beam trawl with a mouth area of 1 6 2 m and a mesh size of 3 mm at
the cod end was used to sample E. pacifica. The benthopelagic population appeared
throughout the year in a bottom layer at 280 m depth, with higher catches in winter
than in summer. A significant catch was made when the thickness of the cold water
layer, with the temperature less than 98C, was more than 55 m in vertical extent.
Observations from submersibles have been made on benthopelagic E. pacifica
over the period from 1994 to 1996 in the sea area near Kinkasan Island, around
38830'N, the central fishing grounds of the species. Additional observations have
been made off Kamaishi, around 39820'N, which is at the northern part of these
fishing grounds (Fig. 3.1.4). Surveys were undertaken using video cameras in daylight during the summer.
For most of the time, the submersibles cruised 0.5 to 1.0 m above the bottom, but
moved away from the bottom in order to determine the numerical density and
dimensions of benthopelagic aggregations. The results are still being analysed
(Endo and Kodama in preparation; videotapes have been prepared from observations made off Kamaishi and have been used for krill analysis, courtesy of Dr. T.
Fujita of National Science Museum, see Stancyk et al., 1998). From these observations, spread over a period of three years, it is evident that benthopelagic E.
pacifica are present over the whole surveyed depth range from 160 to 400 m. The
east to west continuity of some benthopelagic aggregations extended as far as 7 km.
These observations have been limited, however, to two localities, in the sea areas
near Kinkasan and off Kamaishi. Additional observations, made with a beam trawl
by other workers off Ibaraki Prefecture support these findings (Nakamura, 1991;
Ebisawa, 1994).
Observations on the stomach contents of demersal fish over the whole region,
discussed in Chapter 7, indicate that it is very likely that the distribution of
benthopelagic E. pacifica extends about 600 km from Aomori to Ibaraki Prefectures
along the Pacific coast of northern Japan (Fig. 3.1.4).
The water temperature near the bottom varied from year to year and from area to
area, with minima of 28 to 48C in 1994 off Onagawa and 38 to 48C off Kamaishi in
1995 and 1996, and a maximum of 78 to 88C in 1996 off Onagawa. Aggregations of E.
pacifica are rarely encountered during the fishing season, when the water temperature is less than 58C (Odate, 1991; Kodama & Izumi, 1994). Results from a
series of monthly surveys mentioned above over several years off Ibaraki Prefecture
46
Krill: Biology, Ecology and Fisheries
Fig. 3.1.4 Map of northern Japan indicating regions mentioned in text. Submersible observations
were made on benthopelagic population of Euphausia pacifica at two locations shown by black
circles.
Distribution and Standing Stock
47
(Nakamura, 1991; Ebisawa, 1994) indicate that benthopelagic concentrations do not
occur unless the water temperature is less than 98C. The near-bottom water temperature in this region rarely drops below 58C, the lower limit, as noted above, for
such concentrations to occur.
The maximal krill density from these submersible observations was estimated to
be up to 30 000 individuals m±3. As krill did not show parallel orientation, the dense
part of the benthopelagic aggregation corresponds to Mauchline's (1980) definition
of a swarm as opposed to a school in which animals have uniform orientation. The
densest part was restricted to the several metres closest to the bottom.
E. pacifica performs diel vertical migration with daytime depth of 200 to 400 m
and night-time depth of less than 150 m in Sanriku waters (Endo, 1981). Benthopelagic E. pacifica may occur as a result of encountering the sea bed during vertical
migration as suggested for Meganyctiphanes norvegica by Greene et al. (1988). If
this hypothesis is correct, benthopelagic E. pacifica should appear as deep as 400 m;
this corresponds to the upper continental slope. Further offshore, E. pacifica may be
pelagic with a lower population density. Demersal fishes are reported to be most
abundant along the upper slope in Sanriku waters (Fujita et al., 1993), the dominant
species being: walleye pollack, Theragra chalcogramma, and Pacific cod, Gadus
macrocephalus, both of which feed on zooplankton in general, and E. pacifica in
particular. This indicates that benthopelagic krill are probably supporting a very
high abundance of demersal fish in the upper slope region.
Taki et al. (1996) reported the seasonal occurrence of benthopelagic E. pacifica
sampled by a beam-trawl net, which was able to collect plankton from 50 cm to 1 m
above the seabed. The biomass was highest in June (2 g m±2 at 300 m depth) and
lowest in February and April (0.02 g m±2 at 300 m depth). In April, benthopelagic
krill occurred at the depth of 150 m, while they occurred in offshore areas at depths
of 200 to 300 m in other seasons. This seasonal shift of near bottom biomass, estimated from beam-trawl surveys coincided with the distribution of pelagic biomass in
the water column, estimated from net tows (Taki et al., 1996). Stomach contents
analysis of demersal fishes shows a general trend indicating that the importance of
E. pacifica is highest in the period from May to July and lowest in the winter season
(Fujita, 1994). The seasonal shift in benthopelagic krill distribution may be related
to seasonal changes in the temperature in Sanriku waters, with a minimum in spring
around 28C and a maximum in autumn of around 228C at the surface.
Benthopelagic populations are also known for other species of krill such as
Antarctic krill, Euphausia superba (Gutt & Siegel, 1994; Takahashi & Iwami, 1997),
North Atlantic krill, Meganyctiphanes norvegica (Simard et al., 1986; Greene et al.,
1988) and Thysanoessa raschi (Simard et al., 1986). Takahashi & Iwami (1997)
reported that benthopelagic E. superba is distributed down to depths of around 800
m in the vicinity of the South Shetland Islands and constitutes one of the main food
resources for demersal fishes in summer.
Quantitative estimates of the biomass of benthopelagic E. pacifica are needed
over the whole distributional range in the Japanese waters to elucidate better the
48
Krill: Biology, Ecology and Fisheries
importance of the species in the pelagic as well as benthopelagic food webs in
Sanriku and Joban waters.
Abundance estimation using acoustic and spawning survey methods
A feasibility study of species identification and abundance estimation of E. pacifica
was undertaken using a dual-frequency (38 and 120 kHz) echo-sounder in April
1996 (Miyashita et al., l997). E. pacifica form dense aggregations and are the
dominant zooplankters in the fishing season. These characteristics are ideal for the
estimation of standing stock acoustically. Miyashita et al. (1997) adopted the
straight-cylinder model taking account of swimming angle to estimate the acoustic
target strength. This was then used to estimate the size and abundance of krill.
Assuming a swimming angle of 158, the observed value when actively swimming
(Miyashita et al., 1996), the TS difference between the two frequencies (DTS = TS120
7TS38) of E. pacifica of body length 16.5 mm was estimated to be between 10 and
15 dB. Those echoes which fell within this DTS range were assumed to be E. pacifica
(Plate 3, facing p. 182) and thus distinguishable from those echoes due to walleye
pollack, the TS of which, using the swim bladder resonance model, would be less
than 2 dB. The estimated size and numerical density of E. pacifica at one point on
the echogram, for example, were 17.0 mm and 8.3 individuals m±3, respectively.
In April 1997, using the same criteria, Miyashita et al. (1998) estimated the
numerical density of the species along Sanriku and Joban coasts in the depth range
from 150 m to 300 m in April 1997. In this case, the theoretical scattering model
(high-pass sphere model) of Johnson (1977) was used to estimate TS. The highest
density, 9519.5 individuals m±2, was found in the sea area off Fukushima Prefecture
where commercial fishing had been in progress during the survey period. In the
southernmost survey area, where the water temperature had increased rapidly from
98 to 158C, E. pacifica were absent.
Another promising method to assess krill biomass might be the spawning survey method, in which adult biomass is estimated from egg abundance. This
method has been used for sardines and mackerel since the 1940s in Japan and
USA (Watanabe, 1983). Egg abundance is estimated more effectively than is possible with adult krill because eggs do not form dense aggregations and are not
able to undertake any avoidance reaction. Even so, several assumptions must be
made to estimate adult biomass from egg abundance. Taki (1998b) adopted this
method to estimate E. pacifica biomass using samples collected with Norpac nets
hauled vertically from 150 m depth to the surface at 17 stations allocated alongshore in both Sanriku and Joban waters (Fig. 3.1.5). The acoustic survey mentioned above was conducted simultaneously along the cruise track (Miyashita et
al., 1998). The estimated biomass (Wi) of E. pacifica at station i can be obtained
from equation (3.1.1).
Wi ˆ $Ni
…3:1:1†
Distribution and Standing Stock
49
Fig. 3.1.5 Biomass of Euphausia pacifica estimated by acoustic method (a); by the spawning
survey method (solid line) and by net samplings with the improved larva net (broken line) (b); and
water temperature (solid line) and salinity (broken line) at 50 m depth (c); at 17 stations allocated
alongshore from 36850'N to 40820'N (Redrawn from Taki, 1998b.)
where $ is the mean wet weight of E. pacifica, and Ni is the individual number under
unit area of sea surface at station i.
The number of eggs spawned in a day at station i (Pi) is given in equation (3.1.2).
Pi ˆ pi ˆ sNi
m
X
qj j j j
…3:1:2†
jˆk
where is the turnover rate of the time needed for hatching in a day, pi is the
number of eggs per unit surface area (1 m2), s is the survival rate of eggs, qj proportion of individuals of the body length class j, j sex ratio of females, j the number
of spawning episodes in a day, and j the number of eggs spawned by one female per
episode, respectively. If we assume all the spawned eggs hatch in the upper 150 m,
then
Ni ˆ
pi
m
X
s
qj j j j
jˆk
Substituting equation (3.1.3) into equation (3.1.1), we can get
…3:1:3†
50
Krill: Biology, Ecology and Fisheries
Wi ˆ
$pi
m
X
s
qj j j j
jˆk
Taki (1998b) reported high abundance of krill at Stn 28 off Fukushima Prefecture by
this method, which was confirmed by acoustic results by Miyashita et al. (1998) (Fig.
3.1.5). In the spawning survey method, however, two other abundance peaks
appeared in the northern area at Stns 10 and 21, which were not supported by the
acoustic method. The discrepancy may be attributable to the time lag between
spawning and net sampling. The abundance trend estimated by the acoustic method
and the net sampling with the improved larva net showed good coincidence (Fig.
3.1.5).
Both of these methods are likely to provide estimates of standing stock, the
accurate estimation of which, along with estimates of production of E. pacifica are
necessary for the management of this fishery.
References
Ebisawa, Y. (1994) On Euphausia pacifica in the southern Joban area. Gekkan Kaiyo (Kaiyo
Monthly) 26, 255±59 (in Japanese).
Endo, Y. (1981) Ecological Studies on the Euphausiids Occurring in the Sanriku waters with
Special Reference to Their Life History and Aggregated Distribution. PhD. thesis, Tohoku
University, Sendai, 166 pp. (in Japanese with English abstract).
Endo, Y. (1984) Daytime surface swarming of Euphausia pacifica (Crustacea: Euphausiacea)
in the Sanriku coastal waters off northeastern Japan. Mar. Biol. 79, 269±76.
Endo, Y. (1993) Distribution of zooplankton in relation to warm core rings and warm
streamers. Gekkan Kaiyo (Kaiyo Monthly) 25, 94±99 (in Japanese).
Fujita, T. (1994) Importance of planktonic organisms as food of demersal fish population.
Gekkan Kaiyo (Kaiyo Monthly) 26, 236±41 (in Japanese).
Fujita, T., Inada, T. & Ishito, Y. (1993) Density, biomass and community structure of
demersal fishes off the Pacific coast of northeastern Japan. J. Oceanogr. 49, 211±29.
Greene, C.H., Wiebe, P.H., Burczynski, J. & Youngbluth, M.J. (1988) Acoustical detection of
high-density demersal krill layers in the submarine canyons off Georges Bank. Science 241,
359±61.
Gutt, J. & Siegel, V. (1994) Benthopelagic aggregations of krill (Euphausia superba) on the
deeper shelf of the Weddell Sea. Deep-Sea Res. 41, 169±78.
Hanamura, Y., Endo, Y. & Taniguchi, A. (1984) Underwater observations on the surface swarm
of a euphausiid, Euphausia pacifica in Sendai Bay, northeastern Japan. La mer 22, 63±68.
Hattori, H. (1991) Vertical distribution of zooplankton in the warm core off Sanriku (86B)
and adjacent Oyashio water, with special reference to copepods record. Bull. Hokkaido
Natl. Fish. Res. Inst. 55, 59±77 (in Japanese with English abstract).
Hattori, H. (1993) Copepod distribution in a warm-core ring. Gekkan Kaiyo (Kaiyo Monthly)
25, 99±104 (in Japanese).
Iguchi, N., Ikeda, T. & Imamura, A. (1993) Growth and life cycle of a euphausiid crustacean
(Euphausia pacifica Hansen) in Toyama Bay, southern Japan Sea. Bull. Japan Sea Natl.
Distribution and Standing Stock
51
Fish. Res. Inst. 43, 69±81.
Johnson, R.K. (1977) Sound scattering from a fluid sphere revisited. J. Acoust. Soc. Am. 61,
375±77.
Kodama, J. & Izumi, Y. (1994) Factors relevant to the fishing ground formation of Euphausia
pacifica and the relation to the demersal fish resources. Gekkan Kaiyo (Kaiyo Monthly) 26,
228±35 (in Japanese).
Komaki, Y. (1967) On the surface swarming of euphausiid crustaceans. Pacif. Sci. 21, 433±
48.
Komaki, Y. & Matsue, Y. (1958) Ecological studies on the Euphausiacea distributed in the
Japan Sea. Report of the Cooperative Survey of the Warm Tsushima Current and Related
Waters 2, 146±59 (in Japanese).
Mauchline, J. (1980) The biology of mysids and euphausiids. Adv. Mar. Biol. 18, 1±681.
Miyashita, K., Aoki, I. & Inagaki, T. (1996) Swimming behaviour and target strength of isada
krill (Euphausia pacifica). ICES J. Mar. Sci. 53, 303±308.
Miyashita, K., Aoki, I., Seno, K., Taki, K. & Ogishima, T. (1997) Acoustic identification of
isada krill, Euphausia pacifica Hansen, off the Sanriku coast, north-eastern Japan. Fish.
Oceanogr. 6, 266±71.
Miyashita, K., Aoki, I., Asami, N., Mori, H. & Taki, K. (1998) A feasibility study on the
distribution and standing stock of Euphausia pacifica in Sanriku coastal waters in spring.
Rep. Res. Meeting of Krill Res. and Oceanogr. Conditions in the northwest. Pacific, pp. 110±
124 (in Japanese).
Motoda, S. (1957) North Pacific standard plankton net. Inform. Bull. Planktol. Japan 4, 13±15
(in Japanese).
Nakamura, T. (1991) Distribution and fishing ground formation of Euphausia pacifica in the
southern Joban area. Bull. Tohoku Branch Jap. Soc. Sci. Fish. 41, 44±46 (in Japanese).
Nakamura, T. (1992) Recent aspect of krill fishing grounds off Joban±Kashima area in
relation to warming tendency. Bull. Jap. Soc. Fish. Oceanogr. 56, 155±57 (in Japanese).
Nicol, S. & Endo, Y. (1997) Krill fisheries of the world. FAO Fish. Tech. Paper 367, 1±100.
Nishikawa, J., Tsuda, A., Ishigaki, T. & Terazaki, M. (1995) Distribution of euphausiids in the
Kuroshio front and warm water tongue with special reference to the surface aggregation of
Euphausia pacifica. J. Plankt. Res. 17, 611±29.
Odate, K. (1991) Fishery Biology of the Krill, Euphausia pacifica, in the Northeastern Coasts
of Japan. Suisan Kenkyu Sosho (Library of Fisheries Study) No 40, pp. 1±100 (in Japanese
with English abstract).
Ohtsuki, T. (1975) Distribution of euphausiids and physical environment in the south-west
area of the Okhotsk Sea. Monthly Rep. Hokkaido Fish. Exp. St. 32, 1±10 (in Japanese).
Omori, M. (1965) 160 cm opening-closing plankton net-I. Description of the gear. J.
Oceanogr. Soc. Jap. 21, 212±20.
Ponomareva, L. A. (1963) The Euphausiids of the North Pacific, their Distribution and
Ecology, 142 pp., Dokl. Akad. Nauk. SSSR Israel programme for Sci. Transl. 1966.
Sawamoto, S. (1992) Species composition of euphausiids in Suruga Bay, central Japan. Bull.
Inst. Oceanic Res. & Develop., Tokai Univ. 13, 85±96.
Simard, Y., de Ladurantaye, R. & Therriault, J.-C. (1986) Aggregation of euphausiids along a
coastal shelf in an upwelling environment. Mar. Ecol. Prog. Ser. 32, 203±215.
Stancyk, S. E., Fujita, T. & Muir, C. (1998) Predation behavior on swimming organisms by
Ophiura sarsii. In Echinoderms: San Francisco (R. Mooi and M. Telford, eds.), pp. 425±29.
52
Krill: Biology, Ecology and Fisheries
Balkema, Rotterdam.
Suzuki, M. (1986) On the distribution of Euphausia pacifica in the southern Joban and
Kashima-nada areas. Bull. Tohoku Branch Jap. Soc. Sci. Fish. 37, 30±31 (in Japanese).
Takahashi, M. & Iwami, T. (1997) The summer diet of demersal fish at the South Shetland
Islands. Antarct. Sci. 9, 407±413.
Taki, K. (1998a) Horizontal distribution and diel vertical migration of Euphausia pacifica
Hansen in summer in and around a warm-core ring off Sanriku, northwestern Pacific. Bull.
Tohoku Natl. Fish. Res. Inst. 60, 49±61.
Taki, K. (1998b) A feasibility study to estimate standing stock of Euphausia pacifica using a
dual-frequency echosounder and egg production method. T.N.F.R.I. News 53, 4±6 (in
Japanese).
Taki, K. & Ogishima, T. (1997) Distribution of some developmental stages and growth of
Euphausia pacifica Hansen in the northwestern Pacific on the basis of Norpac net samples.
Bull. Tohoku Natl. Fish. Res. Inst. 59, 95±117 (in Japanese with English abstract).
Taki, K., Kotani, Y. & Endo, Y. (1996) Ecological studies of Euphausia pacifica Hansen and
seasonal change of its environment off Onagawa, Miyagi Prefecture III. Distribution and
diel vertical migration. Bull. Tohoku Natl. Fish. Res. Inst. 58, 89±104 (in Japanese with
English abstract).
Terazaki, M. (1980) Surface swarms of a euphausiid Euphausia pacifica in Otsuchi Bay,
northern Japan. Bull. Plankt. Soc. Jap. 27, 19±25.
Terazaki, M. (1992) Horizontal and vertical distribution of chaetognaths in a Kuroshio warm
core ring. Deep-Sea Res. 39, Supplementary Issue No. 1A, S231±S245.
Watanabe, T. (1983) Spawning survey method. In Population Dynamics of Fishery Resources
(T. Ishii, ed.), pp. 9±29. Fisheries Science Series No 46. Koseisha-Koseikaku, Tokyo (in
Japanese).
Watanabe, Y. (1988) Design of surface towing nets and analysis method of sampling data. In
Manual on Sampling and analysis of Eggs and Larvae of Pelagic Fishes. Natl. Res. Inst.
Fish. Sci., pp. 15±22 (in Japanese).
Wiebe, P.H., Morton, A. W., Bradley, A. M. et al. (1985) New developments in the MOCNESS, an apparatus for sampling zooplankton and micronekton. Mar. Biol. 87, 313±23.
Yasuda, I., Okuda, K. & Shimizu, Y. (1996) Distribution and Modification of North Pacific
Intermediate Water in the Kuroshio-Oyashio interfrontal zone. J. Phys. Oceanogr. 26, 448±
65.
Distribution and Standing Stock
3.2
53
Canadian waters
Inigo Everson
Although there has been some commercial interest shown in the krill that are to be
found on both the eastern and western Canadian seaboards, this has been restricted
to two relatively small regions. These are the Nova Scotian shelf and specifically the
St Lawrence estuary on the Atlantic Ocean side and the Vancouver and Queen
Charlotte Islands area on the Pacific Ocean side. Due to their geographical
separation these regions are considered separately.
Pacific Ocean seaboard of Canada
Of the many species of euphausiid known to be present in the area only three,
Euphausia pacifica, ThysannoeÈssa spinifera and T. longipes, figure greatly in ecological studies in the region. Fulton and LeBrasseur (1984) estimated that 90% of
the euphausiid biomass around the Queen Charlotte Islands was composed of these
three species and although they were present on the shelf, the distribution of all
three species extended into deep water. Fulton et al. (1982), when describing results
from winter and spring surveys in 1980 around the Queen Charlotte Islands, found
that the seasonal mean euphausiid biomass was similar in Hecate Strait and Dixon
Entrance but was 2±4 times greater off the west coast of the southern islands in the
group, a region where the continental shelf is very narrow. In the Dixon Entrance,
situated at the northern end of the Queen Charlotte Group T. longipes made up the
greater proportion of the biomass but at the southern end T. spinifera was the
dominant species. These areas of concentration are shown in Fig. 3.2.1.
Further south the trend continues such that E. pacifica and T. spinifera are the
dominant species and account for 90% of the euphausiid biomass around Vancouver Island. Using the results from a series of surveys from 1991 to 1997 (Tanasichuk, 1998a , 1998b) showed the extent of inter-annual variation in standing stock.
During that study, sea temperatures had been anomalously warm in 1992, a period
coincident with an El NinÄo Southern Oscillation (ENSO) event. In that season the
abundance of E. pacifica larvae was found to be six times the level it had been in
1991. In the case of adults, due to the strong recruitment of the 1992 larvae, the
abundance was twice as high in 1992 and 1993 as it had been in 1991. Subsequent to
the ENSO event larval abundance has continued to fluctuate although the abundance of adults returned to the pre-ENSO level.
A slightly different story emerges from an examination of the results for T.
spinifera (Tanasichuk, 1998b). Larval abundance was 2.4 times higher in 1992 than it
had been in 1991 and became low for 1993 and 1994 when it declined to 0.6 and 0.4
respectively of the 1992 value. It was, however, higher again in 1995 at 6.3 and 1996
at 1.9. In spite of this the abundance of adult T. spinifera tells a different story since
there had been a steady decline from 1991 through to 1996 by which time it was only
54
Krill: Biology, Ecology and Fisheries
56 N
Latitude (degrees)
54 N
T.L.
52 N
E. pac.
50 N
T.S.
48 N
132 W
129 W
126 W
123 W
120 W
Longitude (degrees)
Fig. 3.2.1 Coastline of British Columbia showing the latitudinal distribution of the three dominant euphausiids in the region. T.L. = Thysanoessa longipes; T.S. = T. spinifera; E. pac = Euphausia
pacifica.
about 0.14 times the pre-ENSO levels. Tanasichuk (1995) suggests that the reason
for this decline may not be directly related to the environment but might be a result
of changes in predation pressure from fish. Euphausiids figure significantly in the
diet of hake and it appears that these have increased over the same time frame.
Off Vancouver Island in June and August 1986, an acoustic and net sampling
programme found that the macrozooplankton layers were dominated by the
euphausiids E. pacifica and T. spinifera. During both the surveys of this study by
Simard & Mackas (1989) dense aggregations were found in the same two general
regions and shown in Fig. 3.2.2. The first was along the shelf break, where the highest
biomass was observed. A second broader and more diffuse aggregation, covering
about 200 km2, occupied the deeper south-east end of the shelf area with a density of
1±2 g/m2. The scattering layers undertook a consistent diel vertical migration with
daytime depth at the shelf break varying from around 150 to 200 m while on the shelf it
was shallower (~100±150 m) although occasionally extending to the bottom. Simard
& Mackas (1989) noted that these day-depths corresponded to the location of
California Undercurrent waters with a temperature of 6±78C and a salinity of 33.75
ppt. These studies all indicate that while E. pacifica is present throughout most of the
region there is a transition from T. longipes in the north to T. spinifera in the south.
Distribution and Standing Stock
55
52 N
Latitude (degrees)
51 N
50 N
49 N
48 N
128 W
127 W
126 W
125 W
124 W
123 W
122 W
Longitude (degrees)
Fig. 3.2.2 Coastline of British Columbia around Vancouver Island showing region of high
euphausiid biomass, Euphausia pacifica and Thysanoessa spinifera. (From information in Simard &
Mackas, 1989.)
Even though there is an annual total allowable catch (TAC) of 500 t for the
entire British Columbia coast, the main centre of commercial fishing has been
restricted to the Strait of Georgia and concentrated in the Jervis Inlet and adjacent Malaspina Strait (Romaine et al., 1995). This is a small area relative to the
western Canadian seaboard. Jervis inlet, shown in Fig. 13.2, is about 316 km2 and
since it is mostly enclosed short-term changes due to physical transport are
unlikely to be of major significance. A series of acoustic surveys have been
undertaken in the region to provide standing stock estimates. Because the
entrance to the inlet is relatively narrow, with an assumed very low level of interchange with the Strait of Georgia, it has been possible to follow the seasonal
changes at the population level. Two cohorts with modal lengths 9 and 20 mm
were present in June 1994 and the smaller cohort (9 mm) grew over the sampling
period to reach 20 mm by May 1995. The larger (20 mm) cohort seen in June 1994
was not present in samples after October 1994. The loss of this larger cohort at
that time probably explains the decline in biomass up to January 1995 shown in
Fig. 3.2.3.
56
Krill: Biology, Ecology and Fisheries
12000
200 kHz kriging interpolation
200 kHz block averages
100 kHz kriging interpolation
'Sea Truth' net estimates
Inlet biomass (t) wet weight
10000
8000
6000
4000
2000
0
July 94
Sept. 94
Nov. 94
Jan. 95
Mar. 95
May 95
Nov. 95
Month
Fig. 3.2.3 Euphausiid biomass within Jervis Inlet determined from a series of acoustic surveys
supported by net sampling. The pronounced within season variation in biomass is clearly evident.
(With permission from Romaine et al., 1996.)
Eastern Canadian seaboard
The main focus of interest on the Atlantic seaboard of Canada has been in the St
Lawrence estuary, Bay of Fundy and the Scotian shelf where four species are of
significance in this coastal ecosystem, ThysannoeÈssa lomgicaudata, T. raschii, T.
inermis and Meganyctiphanes norvegica.
Sameoto (1982) showed that although T. longicaudata was present on the Scotian
shelf, where the water is relatively colder than offshore, its abundance was much
lower than in the deeper oceanic water at and beyond the shelf break. This largely
confirmed the conclusions of Berkes (1976) who only found this species in the
deeper eastern part of the Gulf of St Lawrence. The distribution of the Scotian shelf
species is indicated in Fig. 3.2.4.
From a study examining surface swarms in the Bay of Fundy using aerial
photography, Nicol (1986) noted that the density of M. norvegica within these
swarms varied from 77.8 to 778 g /m2. Further studies, specifically related to surveys
to estimate standing stock, do not appear to have been made for this locality.
Arising from this a conservative estimate of the density of krill within the bay has
been inferred from estimates coming from the Scotian shelf (DFO, 1996).
The Gulf of St Lawrence is a semi-enclosed sea connected to the Atlantic Ocean
Distribution and Standing Stock
57
54 N
51 N
48 N
45 N
42 N
39 N
72 W
66 W
60 W
54 W
48 W
Fig. 3.2.4 Areas of concentration, shown shaded, of euphausiid biomass on the Nova Scotia shelf.
The main species in the area is Meganyctiphanes norvegica. (DFO, 1996.)
by two straits, the southern Cabot Strait about 450 m deep and the shallower
northeastern Belle-Isle Strait which has a maximum depth of around 70 m. The Gulf
is divided by a network of three channels about 350 m deep which merge in the
Cabot Strait and extend up to the continental slope between the Scotian Shelf and
the Grand Banks. These main features are shown in Fig. 3.2.5.
The circulation in the region is complex and is the result of different forces
operating. The main system is of a two-layer estuarine flow, driven by the large
freshwater discharge of the St Lawrence which strongly influences the mean circulation pattern (Koutitonsky & Bugden, 1991). The freshwater discharge has a
clear annual signature, with a strong peak in Spring following the thaw. Both the
water mass turbidity, which is always higher in the St Lawrence estuary than in the
Gulf, and the phytoplankton production have Spring peaks, which are spatially
modulated (Therriault & Levasseur, 1985). These features which have a major
effect on the local environment appear to be important in controlling the abundance
and spatial structure of euphausiid populations in the region.
Within the Gulf of St Lawrence there have been a series of studies over a number
of years. Analysing results from a series of ten cruises over two years, Berkes (1976,
1977) noted that adults of M. norvegica, T. raschii and T. inermis occurred most
abundantly in the western Gulf while their larvae were present in the Magdalen
Shallows. These three species, along with T. longicaudata, reach reproductive
maturity at one year although T. longicaudata is the only one for which the samples
58
Krill: Biology, Ecology and Fisheries
51 N
Belle-Isle
Strait
50 N
Latitude (degrees)
49 N
48 N
Magdalen
shallows
47 N
Cabot Strait
46 N
45 N
66 W
64 W
62 W
Longitude (degrees)
60 W
58 W
Fig. 3.2.5 Water circulation pattern in the Gulf of St Lawrence shown by arrows (Berkes, 1976)
and main concentration, shown shaded, of Meganyctiphanes norvegica.
indicated that there were never more than two breeding-year classes present in the
population. Spawning in all the three ThysannoeÈssa species takes place in April, a
month that is coincident with the phytoplankton bloom.
A three-year study in the Gulf of St Lawrence estuary, reported by Sameoto
(1976), using a 120 kHz echo-sounder demonstrated the existence of large populations of T. raschii, T. inermis and Meganyctiphanes norvegica. Chlorophyll a concentrations, measured within 5 m of the surface at the same time as the sound
scattering layers were sampled, showed significant correlation with the numerical
density of euphausiids in the area.
In the June 1973 survey M. norvegica was the most abundant species with
aggregations concentrated in the region of the Laurentian channel between the tip
of Gaspe and Anticosti Islands. The same species was also very common and
numerous on the north side of the St Lawrence estuary between Sanuenay River
and Pte des Monts. T. raschii was the dominant species in the area along the north
shore and in the Anticosti Channel as well as along the Gaspe coast whereas
T. inermis was dominant only on the eastern tip of Anticosti Island.
During the September 1973 survey the dominant species was T. raschii which was
Distribution and Standing Stock
59
concentrated in dense scattering layers along Gaspe coast. That same survey showed
there to be small concentrations of M. norvegica in the Laurentian channel at that
time. Continuing the sequence, the December 1973 and May 1974 surveys indicated
that the scattering layers sampled along the coast of Gaspe and in the mouth of the
St Lawrence River west of Pte des Monts contained predominantly T. raschii.
There was a difference in the relative importance of the species from the different
years. Numerically T. raschii was the dominant species over much of the area in 1973
and increased greatly over the period from September to December 1973 as shown
in Table 3.2.1. All three species were concentrated along the Gaspe coast during
May and June, a period when primary production in the area was at a peak.
Table 3.2.1 Ratio of the numerical abundance of each species in each
survey relative to the lowest density for that species.
Date
Ratio
T. raschii
T. inermis
M. norvegica
June 1972
1
7.5
1.5
June 1973
5.8
2.1
1
1.9
Sept 1973
416
1
Dec 1973
165
12
1
May 1974
10
1
1
The results of a study reported by Simard et al. (1986) noted that most of the
euphausiids were aggregated in a patch greater than 100 km in length, but only one
to seven kilometres wide, along the northern edge of the Laurentian Channel.
Within this patch the biomass was greater than 1 g dry wt/m2 and on a scale of
kilometres showed well-defined higher density zones where the biomass was found
to be up to 22.2 g dry wt/m2. Overall the mean density within the patch was around
60 mg dry wt/m3.
In that study euphausiids were found at depths varying from 50 to 175 m with
some of these large concentrations extending down and coming close to the bottom.
The krill were present in two distinct layers with modal depths of 75 and 150 m, the
shallower layer being dominated by T. raschii while the deeper layer contained
mostly M. norvegica. Vertical and horizontal distribution patterns of temperature
and salinity indicated that upwelling was occurring during the survey. The authors
suggest that the aggregations of euphausiids along one edge of the channel resulted
from the interaction of the negative phototaxis of the animals with the upwelling
and mean circulation in the area. It is suggested that the euphausiids were transported towards the north shore by the circulation outlined above. However, they
were not carried up to the surface but remained at a level governed by the depth to
which daylight penetrated underwater. The authors suggest that this resulted in the
60
Krill: Biology, Ecology and Fisheries
formation of a long narrow patch running along the bathymetric contours on the
upwelling side of the Laurentian Channel in the St Lawrence estuary.
Topographic effects on the circulation are also thought to affect the density along
the aggregations so that the channel heads, in particular the head of the Laurentian
channel, are the major sites where the aggregations occur. These same sites are also
known to be the traditional feeding grounds for baleen whales (Simard et al., 1986;
Simard & Lavoie, 1999).
The varying degree of intensity of sampling over the region has made it difficult to
derive an overall estimate of standing stock for the Maritimes Region. Standing
stock estimates of M. norvegica per km2 within two depth strata have been made by
taking the results from the different studies and analysing them together. The two
depth strata selected have been the shelf, where the water is less than 200 m, and
deep water to the continental slope where the depth is between 200 and 1000 m.
Density estimates have been determined acoustically during the spring and summer
as noted above. These results were used along with known areas of distribution
derived from net surveys in order to estimate krill biomass. To allow for uncertainties in the estimation procedure conservative values were taken for the estimates. These were 10 g m±2 as the estimated total density in both the Emerald and
LaHave Basins while a much lower value of 2 g m±2 was used for all other localities
(DFO, 1996). The estimated values are shown in Table 3.2.2.
Table 3.2.2
Krill biomass estimates for the eastern Canadian seaboard. Data from DFO (1996).
Location
Area
(km2)
Emerald and LaHave Basin
5296
Basins in eastern half of Shelf
2399
Fundy Channel and Jordan Basin
Continental shelf edge including Laurentian Channel
(200±1000 m)
Krill biomass
(t)
52 960*
4800**
7714
15 000**
25 406
51 000**
Total Shelf
123 760
* Based on an average biomass of 10 g m±2
** Estimates are based on the conservative assumption of a biomass of 2 g m±2
These results are substantially lower than those given by Runge & Joly (1995)
from a study using net tows in association with information from historical sources.
They estimated the biomass to be 500 000 t for an area of 21 000 km2 in the western
Gulf of St Lawrence. There is no way with the information available to indicate
which of these estimates is the more reliable. They are both estimating standing
stock but using different methods; some difference is unsurprising. What is important is that they demonstrate the level of uncertainty associated with the estimates, a
factor that should be incorporated into any model taking the information forward in
Distribution and Standing Stock
61
order to provide scientific advice for the management of any fishery that might
develop in the region.
References
Berkes, F. (1976) Ecology of euphausiids in the Gulf of St Lawrence. J. Fish. Res. Board Can.
33, 1894±1905.
Berkes, F. (1977) Production of the euphausiid ThysanoeÈssa raschii in the Gulf of St Lawrence. J. Fish. Res. Board Can. 34, 443±46.
DFO (1996) Krill on the Scotian Shelf. DFO Atlantic Fisheries Stock Status Report 96/
106E.
Fulton, J. & LeBrasseur, R. (1984) Euphausiids of the continental shelf and slope of the
Pacific coast of Canada. La mer 22, 268±76.
Fulton, J., Arai, M.N. & Mason, J.C. (1982) Euphasuiids, coelenterates, ctenophores and
other zooplankton from the Canadian Pacific coast ichthyoplankton survey, 1980. Can.
Tech. Rep. Fish. Aquat. Sci., 1125, 75 pp.
Koutitonsky, V.G. & Bugden, G.L. (1991) The physical oceanography of the Gulf of St
Lawrence: a review with emphasis on the synoptic variability of the motion. Can. Spec.
Publ. Fish. Aquat Sci. 113, 57±90.
Nicol, S. (1986) Shape, size and density of daytime surface swarms of the euphausiid Meganyctiphanes norvegica in the Bay of Fundy. J. Plank. Res. 8, 29±39.
Romaine, S.J., Mackas, D.L., Macaulay, M.C. & Saxby, D.J. (1995) Comparisons of repeat
acoustic surveys in Jervis inlet, British Columbia, 1994±1995. In Harvesting Krill: Ecological Impact, Assessment Products and Markets. Fisheries Centre Research Reports, University of British Columbia, Canada. 1995 Vol. 3 No. 3. pp. 48±52.
Runge, J.A. & Joly, P. (1995). Zooplancton (euphausiaceÂs et calanus) de l'estuaire et du golfe
du Saint-Laurent. In Rapport sur l'eÂtat des inverteÂbreÂs en 1994: crustaceÂs et mollusques des
coÃtes du QueÂbec, crevette nordique et zooplancton de l'estuaire et du golfe du Saint-Laurent
(L. Savard, ed.), Chap. 7, pp. 129±37. Rapp. Manus. Can. Sci. halieut. Aquat. 2323.
Sameoto, D.D. (1976) Distribution of sound scattering layers caused by euphausiids and their
relationship to chlorophyll a concentrations in the Gulf of St Lawrence estuary. J. Fish. Res.
Board Can. 33, 681±87.
Sameoto, D.D. (1982) Zooplankton and micronekton abundance in acoustic scattering layers
on the Nova Scotian slope. Can. J. Fish. Aquat. Sci. 39, 760±77.
Simard, Y. & Lavoie, D. (1999) The rich krill aggregation of the Saguenay-St Lawrence
Marine Park: hydroacoustic and geostatistical biomass estimates, structure, variability and
significance for whales. Can. J. Fish. Aquat. Sci. 56, 1182±97.
Simard, Y. & Mackas, D. (1989) Mesoscale aggregations of euphausiid sound scattering layers
on the continental shelf of Vancouver Island. Can. J. Fish. Aquat. Sci. 46, 1238±49.
Simard, Y., de Ladurantaye, R. & Therriault, J.-C. (1986) Aggregation of euphausiids along a
coastal shelf in an upwelling environment. Mar. Ecol. Prog. Ser. 32, 203±215.
Tanasichuk, R.W. (1995) The influence of inter-annual variations in sea temperature on the
population biology and productivity of euphausiids in Barkley Sound, Canada. In Harvesting Krill: Ecological Impact, Assessment Products and Markets. Fisheries Centre
Research Reports, University of British Columbia, Canada. 1995 Vol. 3 No. 3. pp. 62±64.
Tanasichuk, R.W. (1998a) Interannual variations in the population biology and productivity
62
Krill: Biology, Ecology and Fisheries
of Euphausia pacifica in Barkley Sound, Canada, with special reference to the 1992 and
1993 warm ocean years. Mar. Ecol. Prog. Ser. 173, 163±80.
Tanasichuk, R.W. (1998b) Interannual variations in the population biology and productivity
of Thysnaoessa spinifera in Barkley Sound, Canada, with special reference to the 1992 and
1993 warm ocean years. Mar. Ecol. Prog. Ser. 173, 181±95.
Therriault, J.C. & Levasseur, M. (1985) Control of phytoplankton production in the lower St
Lawrence estuary: light and freshwater runoff. Naturaliste Can. 112, 77±96.
Distribution and Standing Stock
3.3
63
The Southern Ocean
Inigo Everson
Large-scale distribution
The circumpolar nature of the Southern Ocean with its dominant West to East
water movement suggests that many planktonic species may be circumpolar in
distribution. That this hypothesis is largely correct was shown by Baker (1954)
studying the zooplankton and Hart (1942) the phytoplankton samples of `Discovery
Investigations' both of whom found that the species composition did not change
markedly around the continent.
There are in the Southern Ocean the following commonly occurring species of
euphausiid crustacean: Euphausia superba, E. vallentini, E. triacantha, E. frigida, E.
crystallorophias and Thysanoessa vicina and T. macrura. Due to the general circulation pattern of the Southern Ocean they are found all around the continent
although they are separated reasonably well latitudinally as is shown in Fig. 3.3.1. Of
the species which commonly occur in the Southern Ocean there are only two, E.
superba and E. crystallorophias, which regularly occur in dense swarms and are
likely to figure in commercial catches.
E. crystallorophias is restricted more or less entirely to the shelf region of the
Antarctic continent. In the south-west Indian Ocean sector, it has been shown by
Hosie (1994) and Hosie & Cochran (1994) that it is predominant in the neritic
zone and conversely that substantial catches of E. superba are unlikely to be
made in localities close to the Antarctic continent. During a recent survey
extending from 808 to 1508 East in the Indian Ocean sector Pauly et al. (1996)
found E. superba in deep oceanic water but as soon as the survey extended on to
the continental shelf the catches were almost entirely composed of E. crystallorophias. In the Pacific sector a recent series of observations reported by Azzali
& Kalinowski (1999) found E. superba extending to around 748 South in the
eastern part of the Ross Sea and only south of that latitude was E. crystallorophias the dominant species.
The high latitude neritic distribution for E. crystallorophias is also typical of the
Atlantic sector (Mauchline & Fisher 1969) although there have been records away
from the Antarctic coastal region. This species has been reported from the caldera
of Deception Island, South Shetlands by Everson (1987), another shallow locality,
while Brierley & Brandon (1999) reported a concentration of E. crystallorophias
from deep oceanic water in the Scotia Sea which appeared to be within a tongue of
cold continental shelf water.
By far the most important species commercially in the Southern Ocean is the
Antarctic krill, E. superba, and the remainder of this part of the chapter will be
restricted to its consideration. The broad scale distribution of E. superba has been
described by Marr (1962) from a large series of net hauls during the `Discovery
64
Krill: Biology, Ecology and Fisheries
Fig. 3.3.1 Latitudinal distribution of the species of the genus Euphausia in the Southern Ocean.
The shaded portion represents the main range, the open portion the extreme range. With permission from Cambridge University Press. (After John, 1937.)
Investigations.' Although there have been many large-scale surveys subsequently,
that picture of the general distribution extending from the high Antarctic continental shelf north as far as the Antarctic Polar Front Zone (APFZ) and shown in
Fig. 3.3.2. is still accepted. It was this distribution which was behind the designation
of the management area for the Commission for the Conservation of Antarctic
Marine Living Resources (CCAMLR). The CCAMLR area, shown in Fig. 12.1, was
designed to cover the distribution of all species harvested within the Southern
Ocean and for these the APFZ provides a very convenient northern limit as proposed by Everson (1977). The initial plan, with the exception of the part between
558 and 608 South and 508 to 608 West which was excluded for political reasons, in
the Atlantic sector was accepted. This locality, within FAO Division 41.3.2, is the
only area outside of the CCAMLR region from which Antarctic krill catches have
been reported (Anon, 1998).
Distribution and Standing Stock
65
30 W
0
60 W
30 E
90 W
120 W
60 E
80 S
70 S
90 E
60 S
150 W
50 S
180 W
40 S
120 E
150 E
Fig. 3.3.2 Large-scale distribution of krill. Based on surveys and commercial fishing activity.
(Based on Marr, 1962; Everson & Miller, 1994; Nicol & Endo, 1997.)
Large-scale estimation of standing stock
With a distribution extending over much of the 36 million square kilometres which
make up the Southern Ocean, estimation of the standing stock of E. superba presents enormous logistical difficulties. A broad-scale survey with transects spaced
even as close as 30 mile intervals would take about 30 vessels a month to complete,
clearly an unrealistic commitment. Added to this is the logistical problem caused by
a significant part of the area being covered by sea-ice. It is no surprise therefore that
there has never been a single synoptic survey for this species.
Three large-scale surveys have been undertaken. The first to estimate the
standing stock of krill in the Southern Ocean, formed a major part of the Biological
Investigations of Marine Antarctic Systems and Stocks (BIOMASS) programme
and as the First International BIOMASS EXperiment (FIBEX) covered a large part
of the Southwest Atlantic and Western Indian Ocean sectors. This was a major
66
Krill: Biology, Ecology and Fisheries
undertaking involving 11 vessels from 10 nations (Hempel, 1983) and involved a
flexible survey design to take account of time lost due to adverse weather (Everson
et al., 1989). That design was developed further and used for other acoustic surveys
(Jolly & Hampton, 1990). The results from the survey are summarised in Table
3.3.1.
Table 3.3.1 Krill (Euphausia superba) standing stock estimates from the FIBEX
survey which took place between 16 January and 12 March 1981. Data from
Trathan et al., 1992.
CCAMLR or FAO
Area of subarea
Area surveyed
(km2)
Standing stock
million tonnes
CV
(%)
75 000
3.66
29.6
48.1
283 000
10.54
35.0
48.2
242 000
15.61
22.2
48.3
25 000
1.51
37.9
48.6
576 000
4.63
22.9
1 711 000
3.93
32.0
41
58.4.2
During January and February 2000 there was a further survey in the Atlantic
sector to estimate standing stock over a larger area than had been covered by the
FIBEX survey. All four participating vessels were equipped with identical multifrequency echosounder and data management systems. Preliminary analyses of the
data indicate that the estimated standing stock of 44.3 million tonnes (SC-CAMLR
2000) is similar to that estimated from the FIBEX results.
The 1996 survey in the Southeast Indian Ocean sector by the Australian Antarctic
Division was undertaken by a single vessel over a period of two months and
extended from 808 to 1508 East (Pauly et al., 1996). In order to cover the area
designated for the survey, the transects were widely spaced, depending on the
sampling regime along a particular transect the spacing was 150 or 80 nautical miles.
Density was not uniform over the area and krill were more abundant in the west of
the surveyed area, between 808 and 1208 East. The results are summarised in Table
3.3.2.
Although on a much smaller scale, the surveys undertaken by Italian scientists
working in the Ross Sea provide further insight into the distribution of krill (Azzali
& Kalinowski, 1999). These surveys in 1989±90 and in 1994 demonstrated that
E. superba were present as far as 768 South in an area previously thought to contain
few krill (Marr, 1962).
These large-scale acoustic surveys although covering less than half of the
Southern Ocean have provided confirmation of the pattern of distribution determined from nets in these areas. The swarming nature of krill, discussed by Watkins
in Chapter 4, means that the distribution is decidedly non-uniform. This component
Distribution and Standing Stock
67
Table 3.3.2 Results of an Australian krill survey in the Indian Ocean sector of the
Southern Ocean from 29th January to 21st March 1996. Data from Pauly et al.
(2000).
Longitudinal range
2
Area (million km )
80±1158E
115±1508E
2.34
2.34
2
10.6
6.7
Weight density (g/m )
6.7
4.2
Number density (#/m )
2
Standing stock (million of tonnes)
Coefficient of Variation (%)
3.0
19
1.8
30
of the spatial distribution must also be taken into account in order to gain an
understanding of the system.
Small-scale distribution
Turning to a smaller scale of distribution, of the order of one to ten kilometres, we
need to look at situations where intensive sampling has been taking place within
limited areas and a good starting point for this is through an analysis of commercial
fishing activity.
Several studies have been undertaken which look at the spatial distribution of
krill fishing. From an analysis of data in 10-day intervals at a scale of half a degree of
latitude by one degree of longitude Everson & Goss (1992) showed that in the
Atlantic sector fishing tended to be restricted to the coastal or shelf region. On a
finer scale, haul by haul data have been analysed by Endo & Ichii (1989), Ichii et al.
(1992), Kasatkina & Latogursky (1991), Vagin et al. (1992), Sushin (1998) and
Trathan et al. (1998) and all these studies indicate that in the Atlantic sector the
commercial fishery concentrates on the continental shelf close to the shelf break
around the main island groups. Elsewhere there have been few studies although in
the Indian Ocean sector similar analyses have shown that the fishery there is concentrated at the shelf break (Ichii, 1990).
There are two reasons why the fishery concentrates in coastal regions in the
Atlantic sector. The obvious one is that this is where the densest concentrations of
krill are likely to be found. The other explanation is that the coastal region provides
shelter for fishing vessels in the event of adverse weather. Additionally, in the case
of the South Shetlands, the fishing grounds are in closer proximity to port facilities
in South America than anywhere else in the Southern Ocean. These points are
discussed further by Ichii, in Chapter 9.
Examination of the distribution of fishing activity might lead to the conclusion
that the shelf area and shelf break zone are the only major sites of krill abundance.
In terms of abundance estimation it would be very convenient if that were the case
since with much lower concentrations offshore this would greatly reduce the area of
68
Krill: Biology, Ecology and Fisheries
a high density stratum for abundance estimation surveys. Consideration needs to be
given to this situation by assessing what proportion of the krill stock is present in the
oceanic deep water between the main shelf regions. This in turn raises further
questions concerning the relationship between the krill that are present around one
island group with those present in locations further `downstream.'
Interactions between areas
Surface water in the West Wind drift flows in a manner analogous to a conveyer belt
travelling from west to east on which krill can be carried over the deep ocean. Everson
& Murphy (1987) demonstrated that krill swarms can be carried within the surface
current in the coastal waters of the Bransfield Strait. From an analysis of the locations
of sequential hauls by commercial fishing vessels Endo & Ichii (1989) demonstrated
that over a period of a few days individual trawlers were targeting on the same krill
concentration as it was carried along the north coast of the South Shetland Islands in
the eastward flowing current. There is therefore good evidence that krill are likely to
be carried on the ocean currents and probably at the same speed.
Extending this idea to the Scotia Sea and using the Fine Resolution Antarctic
Model (FRAM) to determine the main flow fields, Murphy et al. (1998) and Hofmann et al. (1998) demonstrated that krill from around Elephant Island could be
carried across the Scotia Sea to arrive off South Georgia several months later. Thus
although the largest concentrations of krill may be found on the shelf and at the
shelf break it is highly likely that these concentrations are being carried at some time
over the deep ocean between the island groups. Looked at it in this light it appears
that the circumpolar current is effectively a conveyer belt on which the krill are
carried. In this simple conceptual framework it is right to expect krill concentrations
to be found almost anywhere within the Antarctic Circum-Polar Current (ACC).
Commercial fishing, as already noted, tends to be concentrated in a few restricted
localities. At the start of the fishing season at South Georgia the fleets generally
begin operations to the north-east of Cumberland Bay (Trathan et al., 1998), while
in the South Shetland Islands the fleet appears to be operating most regularly to the
north of King George Island (Endo & Ichii, 1989). Although krill concentrations
have been encountered during research surveys in the deep ocean their location
appears to be far less predictable than those found on the shelf or shelf-break. Thus
although the commercial fleets provide some insight into the location of krill
patches, this is only part of the story and because the current krill catch is very much
less than the estimated production that proportion on the shelf may be only a small
part of the overall standing stock.
Several studies have looked at the mesoscale distribution of krill and a good
example is that reported by Sushin & Shulgovsky (1999). The results, summarised
on a subarea scale, are shown in Table 3.3.3. These show that in the three study
periods there was reasonable consistency between the South Shetlands and South
Orkneys but that the pattern at South Georgia was totally different. Various authors
Distribution and Standing Stock
69
Table 3.3.3 Krill standing stock estimates from research surveys in the Southwest Atlantic sector.
The values are the mean (g/1000 m3) along with upper and lower boundary of the 95% confidence
intervals for the estimates. (Data from Sushin & Shulgovsky, 1999.)
Dates of surveys
Subarea 48.1
(South Shetlands)
Subarea 48.2
(South Orkneys)
Subarea 48.3
(South Georgia)
27 Jan±16 Mar 1984
119
(38±231)
105
(26±205)
1.5
(0.15±3.14)
11 Oct±14 Dec 1984
177
(56±95)
131
(65±208)
0.6
(0.29±0.88)
20 Jan±9 Mar 1988
48
(15±91)
71
(26±132)
147.5
(78±238)
have noted this phenomenon and concluded that it is a result of the oceanic circulation pattern across the Scotia Sea (e.g. Mackintosh, 1973; Priddle et al., 1988,
Maslennikov & Solyankin, 1988; and Sushin & Shulgovsky, 1999). Proximity to each
other and their location within the circumpolar current means that links between
the South Shetlands and South Orkneys are likely to be strong. By contrast South
Georgia lying to the northern side of the ACC is likely to be more greatly affected
by any latitudinal variation of the main flow as it passes through the Drake Passage.
As well as highlighting differences at the statistical subarea scale (subareas are
defined by Miller & Agnew, Chapter 12), the study by Sushin & Shulgovsky (1999)
also demonstrated differences within these localities. One feature observed was the
existence of zones of high biomass to the west of about 408 West. These appear to be
similar in location to some favoured fishing grounds although there are important
differences particularly in the Bransfield Strait. The authors note a much greater
degree of variation in the eastern part of the area which they attribute to variation in
the geostrophic currents within the ACC already noted.
Arising from this we can conclude that a main factor controlling the distribution
of krill is the broad-scale geostrophic circulation. This brings the krill to the shelf
regions within which there are further oceanic components affecting the distribution. In this region we have again components of scale which need to be considered,
a shelf and shelf-break scale covering perhaps 50 to 100 km and the finer-scale
effects which appear to be topographically controlled.
Models of local krill distribution
From an analysis of large numbers of net hauls in the vicinity of the South Shetlands,
Siegel (1988), Siegel et al. (1997, 1998) noted a seasonal pattern of distribution of
juvenile, subadult and adult stages, shown diagrammatically in Fig. 3.3.3. The initial
concepts are set out in Siegel (1988). In summer, during the spawning season, there
is a spatial separation of developmental stages along the Antarctic Peninsula, from
the inner shelf where the juveniles are found, to oceanic waters where the adults are
70
Krill: Biology, Ecology and Fisheries
Possible
spawning period
Adult
Subadult
Krill abundance (arbitrary unit)
Juvenile
Continental slope
Shelf
Distance from Continent: Shelf -Oceanic
Northern distribution limit
Jul
Sep
Nov
Jan
Mar
May
Jul
Season
Fig. 3.3.3 Generalised picture of the seasonal fluctuations in the abundance and distribution of
krill in waters of the Antarctic Peninsula and the spatial succession of developmental stages
between coastal and oceanic waters. (With permission from Siegel et al., 1997.)
present. Overlying this is a temporal succession of abundance from early spring
when juveniles predominate through to the summer when all stages are present.
Siegel and his colleagues develop this concept by postulating an offshore migration
in summer and onshore migration during autumn and winter which is represented
diagrammatically in Fig. 3.3.4. Support for these ideas comes from the study by Ichii
et al. (1998) working in the vicinity of the South Shetland Island. They found that
over the period from early to mid-summer the krill density increased and showed
distinct offshore±onshore differences in abundance and maturity stages. In midsummer the krill density was low, 8 g m±2, in the oceanic region and higher, 36 g m±2
in the frontal zone on the slope. The highest values, 131 g m±2, occurred along the
shelf break in the inshore zone.
The annual cycle proposed by Siegel came as a result of the analysis of data from
several calender months over a number of years. Working in an area further south of
Elephant Island in the Palmer Long Term Ecological Research Programme
(LTER), Quetin et al. (1996) undertook an extensive sampling programme over a
number of seasons. Their results from winter sampling led them to develop
hypotheses linking krill distribution and production to the extent of sea-ice cover.
They suggest that when the winter sea-ice is extensive the ice-associated biota
provide favourable feeding conditions for juvenile krill with the result that standing
stock in the following season will be higher. A similar conclusion was reached by
Kawaguchi & Satake (1994) who noted that heavy sea-ice in the austral winter leads
Distribution and Standing Stock
60
W 78
72
A
S
J
A
64
W
56
A
A
A
60
S
62
S
S
S
64
S
S
66
66
A
A
S
68
W
W
60
S
Spring/
Summer
78
72
68
64
60
56
78
72
68
64
60
56
S
A
S
J
L
62
?
66
J
68
W
78
72
68
W
W
60
L
L
A
A
S
J
S
J
A
S
L
L
SJ
A
S
A SJ J
A
JS
66
A
S
68
62
64
Autumn/
Winter
A
S
? J
S
S
L
Geostrophic current
Krill migration
Adult
Subadult
Juvenile
Larvae
L
64
(b)
60
S
A
(a)
64
Geostrophic current
Krill migration
Adult
Subadult
Juvenile
A
S
62
68
71
64
60
56
S
68
W
Fig. 3.3.4 Seasonal distribution pattern of juvenile, subadult and adult krill developmental stages
off the Antarctic Peninsula. (a) Krill migration into oceanic waters during the spring/summer
spawning season. (b) Post-spawning movement into the neritic zone. (With permission from Siegel
et al. 1988.)
to a relatively high abundance of small krill in the population in the following spring.
This is in line with Quetin et al. (1996) who suggest that years of greatest sea-ice
extent will result in a higher reproductive output, as measured by egg production, in
the following season. These two criteria, however, are not always in phase because
the recruitment potential reflects environmental conditions prior to the spawning
season and winter survival, a result of high production during the season, reflects
conditions in the following season (Ross & Quetin, 1991).
All of these studies have been developed as a result of extensive research activity
taking place in one sector of the Southern Ocean. The types of model developed
72
Krill: Biology, Ecology and Fisheries
there might also have application in other regions such as in the Indian Ocean sector
where krill have been the subject of commercial fishing at the shelf break (Endo &
Ichii, 1989).
The Siegel and the Quetin et al. models incorporate information from direct
observations on the krill and relate them to the water circulation. They do not
explicitly take account of predation on krill during the summer period, a topic
discussed by Everson, Chapter 8. Thus a reduction in standing stock towards the end
of the summer period could be due to predation on krill by land-based predators
during the breeding season. If that were the case then the simpler model of the
conveyer outlined earlier in this chapter would apply. Without such detailed
information it is impossible to develop these ideas further.
The data derived from the series of surveys in the vicinity of Elephant Island
highlight a further point with regard to the interpretation of the data. Sampling
was undertaken by two methods and the results presented separately. These are
brought together in Table 3.3.4. When comparing these results there are some
key points that need to be borne in mind. The first is that the density units are
different, acoustics are presented relative to sea surface area whereas the net
hauls are presented as relative to the volume of water filtered. Providing both
methods are sampling the same depth horizon this will not affect the trend, only
the absolute values. The second point is that there is a different pattern in the
seasonal values by the two methods. The reason for this difference is because the
net haul data have been restricted to observations made during January and February, the period during which, according to the Siegel hypothesis, the abundance
should be at its greatest. The acoustic data cover a much wider period of the
summer season.
Small scale distribution
Moving now to a finer scale of the order of kilometres there are krill patches. A
good example is the `super-swarm' described by Mathisen & Macaulay (1983) which
was present on the shelf close to Elephant Island, South Shetlands. This large krill
patch persisted for several weeks and was targeted by commercial fishing vessels.
Similar patches have been noted by other workers (see Watkins, Chapter 4)
although in these cases no commercial fishing was taking place in the vicinity. A
detailed discussion of these features is outside the scope of the present chapter but
the key point to be noted is that they persisted for days or even weeks. In order for
this to happen the localities must have been sites where the krill, for some reason,
became concentrated. This may appear to be contrary to the conveyer belt analogy
developed for the deep ocean situation described above.
In their study in the Bransfield Strait, Everson & Murphy (1987) noted that in the
strongly flowing coastal current to the south of King George Island krill swarms
appeared to be moving with the current. Close by in the channel between King
George and Elephant Islands the water movements were more confused and a clear
Distribution and Standing Stock
73
Table 3.3.4 Estimates of krill density in the vicinity of Elephant Island from
acoustics (Hewitt & Demer, 1994) and net hauls (Siegel et al., 1997).
Acoustic
Date
Mar 1981
Net hauls
Density (g/m2)
68.5
Season
Density (N/1000 m3)
1977/78
348.3
1980/81
161.4
1981/82
324.87
1982/83
276.7
241.7
Nov 1983
13.4
Mar 1984
22.5
1983/84
Nov 1984
Mar 1985
63.7
8.4
1984/85
85.85
Jan 1987
57.2
Jan 1988
41.6
1987/88
27.64
Feb 1989
82.4
1988/89
79.84
Early Jan 1990
Late Jan
Early Feb
Late Feb
17.1
46.9
78.4
90.9
1989/90
15.39
Late Jan 1991
Late Feb
23.8
28.8
1990/91
6.46
Late Jan 1992
Early Mar
61.2
29.6
1991/92
19.86
134.5
88.2
1992/93
29.17
1993/94
32.96
1994/95
73.31
1995/96
119.97
1996/97
212.51
Jan 1993
Feb
pattern was not present; the krill swarms in that region did not appear to be moving
nearly as fast. That study used the set and drift information derived from successive
satellite passes, recorded by a transit satellite navigator, at intervals of an hour or
two. In the study it was assumed that the vessel was being set by an amount indicative of the surface current.
Recently the use of Acoustic Doppler Current Profilers (ADCP) has meant that
information at a much finer horizontal resolution is available and furthermore that
the movements through the water column can also be quantified (Brandon et al.,
2000). Although there have not been studies whereby individual patches have been
monitored for a period of time using ADCP the information on water circulation on
74
Krill: Biology, Ecology and Fisheries
the shelf indicates that the local eddying effects are such that krill patches could be
retained at a location for quite some time.
Further evidence for this can be inferred from the locations of commercial fishing
around South Georgia. Analyses of haul by haul data have demonstrated that
fishing operations commence at more or less the same location each season
(Murphy et al., 1997; Trathan et al., 1998). This demonstrates that some locations,
such as the shelf break north-east of Cumberland Bay, are ones where good concentrations are likely to be encountered regularly.
While this information is supportive of the general picture of the krill, with
localised areas of concentration, this needs to be incorporated into a model which
includes population dynamics. Thus the age-density of the krill present at a given
location must have arisen as a result of population recruitment, growth and mortality. Where part of the population is sampled in one location the age composition
should reflect these demographic changes at sites sampled downstream at a later
date. Watkins (1999) has shown that in some seasons this is the case but there are,
even so, some disparities which remain unexplained. Clearly there is more scope for
research into understanding the distribution of E. superba at all time and space
scales.
Application of information on Southern Ocean krill distribution
Management units
Having considered the distribution of E. superba in the Southern Ocean we can now
consider applications of this information. Because of its central role in the Southern
Ocean food web knowledge of the ecology of krill is central to understanding the
dynamic functioning of the ecosystem. As an extension of this concept, it is
important in the development of advice for management of the fishery.
The foregoing part of this chapter has shown the broad geographical range of
E. superba from the continental shelf zone northwards as far as the APFZ.
Locally there are concentrations some of which occur in predictable localities but
others are dependent on oceanic circulation. At the Southern Ocean scale we
have a very large management unit which due to its enormous area is of limited
meaningful value for management. Subdivision of the area in some form is therefore essential and it is advisable that any such breakdown should have a strong
ecological justification.
The CCAMLR region is divided into three major Areas, 48 the Atlantic Ocean
sector, 58 Indian Ocean sector and 88 the Pacific Ocean sector. This division was
made by extending the FAO Statistical Areas southwards into the Southern Ocean.
While such a division outside the Southern Ocean can be seen to have an ecological
justification, the circumpolar continuity of the Southern Ocean means that such a
division is arbitrary in nature and of minor ecological significance.
Within these areas there have been further subdivisions most of which were
Distribution and Standing Stock
75
designated so as to separate areas of continental shelf which harboured different finfish stocks (Everson, 1977). Due to the localisation of the commercial krill fishery on
the shelf and at the shelf break these same subareas are also reasonable descriptors
of the limits of the fishery. However the circumpolar current means that there are
clear and recognisable interactions between these subareas.
Although the Southern Ocean circulation is basically circumpolar it is subject to a
large degree of variation due to internal variation, atmospheric circulation and
bottom topography. Arising from this there are regions of gyres, some of which have
been thought of as containing local populations. With the recent developments in
molecular biology it has been considered possible to identify such `populations'.
Unfortunately the results of such studies have been unable to determine the extent
of isolation or mixing. In the absence of a clear molecular biology signal alternative
approaches have been considered.
Krill flux
Central to these have been a consideration of the oceanographic circulation as a key
to the transport of krill to and through a region. This concept, referred to as `krill
flux', is being actively considered within CCAMLR.
Studies of krill flux can be made at a variety of different scales. At the large
oceanic scale, where the flow fields are well described by the geostrophic circulation, the situation can be modelled as an extension of the work described by Murphy
et al. (1998)and Hofmann et al. (1998). Greater problems arise in the vicinity of the
shelf break and on the shelf where the circulation becomes less predictable. These
same regions, as outlined above, are the sites where the denser krill patches have
been found and also where commercial fishing takes place.
To address this problem consideration has been given to dividing the region into
small rectangles (Anon. 1994). Estimation of the flow field across and the krill
density along the rectangle boundaries can be used to provide an indication of the
krill flux into and out of the rectangle. The theoretical basis for such a model to
estimate the flux between two adjacent boxes shown in Fig. 3.3.5 was developed in
Anon. (1994 Attachment D) and is set out below. For this the residence time in box
1 (r1) is given by V1/fO1 and that for box 2 (r2) by V2/(f12 + fO2). Note that the
subscript convention used is `O' Outside with the numbers referring to individual
boxes. The overall residence time R is:
Rˆ
V1
fO1
fO1
fO1 ‡ fO2
‡
V2
f12 ‡ fO2
f12 ‡ fO2
fO1 ‡ fO2
This equation can be reorganised since the first and third factors are the residence
times r1 and r2 described above. The second and fourth factors, termed `pooling
factors' in Anon. (1994) and designated w1 and w2, lead to a simplified version of the
equation as:
76
Krill: Biology, Ecology and Fisheries
1-BOX SYSTEM ± Example
fO1
?
1
?
V1
f.1O
V1 = volume (e.g., water volume) in box 1 (e.g., km3)
fO1 = input from `outside' into box 1 (e.g., in km3/day)
f10 = outflow from box 1 to the `outside' (e.g., in km3/day)
The subscript `O' refers to `outside'
fO1
T1 = turnover for box 1 =
V1
V1
r1 = residence time in box 1 =
(e.g., in days)
fO1
2-BOX SYSTEM ± Example
;fO2
1
fO1
?
f12 2
V1
?
V2
?
f2O
Fig. 3.3.5 Schematic representation of the krill flux model. With permission from the SCCAMLK. (Anon., 1994.)
R ˆ r1 w1 ‡ r2 w2
This can then be taken to the generalised case for a large number of boxes:
Rˆ
N
X
ri w i
iˆ1
Estimates of residence times have been made using the data from FRAM and the
FIBEX study and these fall into line with expectation. Those analyses were
undertaken to demonstrate the general validity of the process. They are not
reproduced here because for them to have any real purpose it is important that
the flux estimates are calculated so as to reflect a specific area of interaction.
Such areas might be for management purposes but equally can reflect the overlap
areas of commercial fishing and predator foraging. The importance of this information can be seen when the implications of the management issues that are
raised in Chapters 8 and 12 are considered. The ecosystem approach to management requires that consideration be given to the target species, in this case krill,
Distribution and Standing Stock
77
and the species dependent upon it. Thus the management units need to cater for
the spatio-temporal variation in krill, water circulation as well as those of the
dependent species.
References
Anon. (1994) Report of the workshop on evaluating krill flux factors. Report of the thirteenth
meeting of the Scientific Committee for the Conservation of Antarctic Marine Living
Resources. SC-CAMLR-XIII, Annex 5, Appendix D.
Anon. (1998) Report of the seventeenth meeting of the Scientific Committee for the Conservation of Antarctic Marine Living Resources. SC-CAMLR-XVII.
Azzali, M. & Kalinowski, J. (1999) Spatial and temporal distribution of krill Euphausia
superba biomass in the Ross Sea (1989±1990 and 1994). In Ross Sea Ecology Italiantartide
Expeditions (1987±1995) (F.M. Faranda, L. Guillielmo & A. Ianora, eds), Chapter 31,
pp. 433±55. Springer-Verlag, Berlin.
Baker, A. de C. (1954) The circumpolar continuity of Antarctic plankton species. Discovery
Rep. 27, 201±218.
Brandon, M., Murphy, E.J., Trathan, P.N. & Bone, D.G. (2000) Physical oceanographic
conditions to the northwest of the sub-Antarctic Island of South Georgia. J. Geophys. Res.
(In press.)
Brierley, A.S. & Brandon, M.A. (1999) Potential for long-distance dispersal of Euphausia
crystallorophias in fast current jets. Mar. Biol. 135, 77±82.
Endo, Y. & Ichii, T. (1989) CPUEs body length and greenness of Antarctic krill during 1987/
88 season in the fishing ground north of Livingston Island. Scientific Committee for the
Conservation of Antarctic Marine Living Resources, Selected Scientific Papers. SCCAMLR-SSP/6, pp. 323±45.
Everson, I. (1977) The Southern Ocean. The living resources of the Southern Ocean. FAO
Southern Ocean Fisheries Survey Programme GLO/SO/77/1, 156 pp.
Everson, I. (1987) Some aspects of the small scale distribution of Euphausia crystallorophias.
Polar Biol 8, 9±15.
Everson, I. & Goss, C. (1992) Krill fishing activity in the southwest Atlantic. Antarctic Sci. 3,
351±58.
Everson, I. & Murphy, E. J. (1987) Mesoscale variability in the distribution of krill Euphausia
superba. Mar. Ecol. Prog. Ser. 40, 53±60.
Everson, I., Hampton, I. & Jolly, G.M. (1989) Survey design to estimate krill abundance
during FIBEX. Scientific Committee for the Conservation of Antarctic Marine Living
Resources, Selected Scientific Papers. SC-CAMLR-SSP/5, pp. 253±63.
Hart, T.J. (1942) Phytoplankton periodicity in Antarctic surface waters. Discovery Rep. 21,
261±356.
Hempel, G. (1983) FIBEX ± An international survey in the Southern Ocean: Review and
outlook. Memoirs of the National Institute of Polar Research, Special Issue No 27, 1±15.
Hewitt, R.P. & Demer, D.A. (1994) Acoustic estimates of krill biomass in the Elephant Island
Area: 1981±1993. CCAMLR Sci. 1, 1±6.
Hofmann, E., Klinck, J.M., Locarnini, R.A., Fach, B. & Murphy, E. (1998) Krill transport in
the Scotia Sea and environs. Antarctic Sci. 10, 406±415.
Hosie, G.W. (1994) The macrozooplankton communities in the Prydz Bay region Antarctic.
78
Krill: Biology, Ecology and Fisheries
In Southern Ocean Ecology: The BIOMASS Perspective (S.Z. El-Sayed, ed.), pp. 93±123.
Cambridge University Press, Cambridge.
Hosie, G.W. & Cochran, T.G. (1994) Mesoscale distribution patterns of macrozooplankton
communities in Prydz Bay, Antarctica ± January to February 1991. Mar. Ecol. Prog. Ser.
106, 21±39.
Ichii, T. (1990) Distribution of Antarctic krill concentrations exploited by Japanese krill
trawlers and minke whales. In Proceedings of the National Institute of Polar Research
Symposium in Polar Biology 3, 36±56.
Ichii, T., Ishii, H. & Naganobu, M. (1992) Abundance size and maturity of krill (Euphausia
superba) in the krill fishing ground of Subarea 48.1 during 1990/91 austral summer. Scientific Committee for the Conservation of Antarctic Marine Living Resources, Selected
Scientific Papers. SC-CAMLR-SSP/9, pp. 183±99.
Ichii, T., Katayama, K., Obitsu, N., Ishii, H. & Naganobu, M. (1998). Occurrence of Antarctic
krill (Euphausia superba) concentrations in the vicinity of the South Shetland Islands:
relationship to environmental parameters. Deep-Sea Res. I 45, 1235±62.
John, D.D. (1937) The southern species of the genus Euphausia. Discovery Rep. 14, 195±324.
Jolly, G.M. & Hampton, I. (1990) Some problems in the statistical design and analysis of
acoustic surveys to asses fish biomass. Rapp. P.-v. ReÂun. Cons. Perm. Int. Explor. Mer 189,
415±20.
Kasatkina, S. M. & Latogursky, V.I. (1991) Distribution characteristics of krill aggregations in
the fishing ground off Coronation Island in the 1989/90 season. Scientific Committee for the
Conservation of Antarctic Marine Living Resources, Selected Scientific Papers. SCCAMLR-SSP/7, pp. 49±73.
Kawaguchi, S. & Satake, M. (1994) Relationship between recruitment of the Antarctic krill
and the degree of ice cover near the South Shetland Islands. Fish. Sci. 60, 123±24.
Mackintosh, N.A. (1973) Distribution of post larval krill in the Antarctic. Discovery Rep. 36,
95±156.
Marr, J.W.S. (1962) The natural history and geography of the Antarctic krill (Euphausia
superba Dana). Discovery Rep. 32, 33±464.
Maslennikov, V.V. & Solyankin, E.V. (1988) Patterns of fluctuations in the hydrological
conditions of the Antarctic and their effect on the distribution of Antarctic krill. In Antarctic Ocean and Resources Variability D. Sahrhage, ed.), pp. 209±213. Springer-Verlag,
Berlin.
Mathisen, O.A. & Macaulay, M.C. (1983) The morphological features of a super swarm of
krill, Euphausia superba. Memoirs of the National Institute of Polar Research, Special Issue
No 27, 153±64.
Mauchline, J.R. & Fisher, L.R. (1969) The biology of euphausiids. Adv. Mar. Biol. 7, 1±454.
Murphy, E.J., Everson, I., Parkes, G. & Daunt, F.H.J. (1997) Detailed distribution of krill
fishing around South Georgia. CCAMLR Sci. 4, 1±17.
Murphy, E.J., Watkins, J.L., Reid, K., et al. (1998) Interannual variability of the South
Georgia marine ecosystem: biological and physical sources of variation in the abundance of
krill. Fish. Oceanogr. 7, 381±90.
Pauly, T., Higginbottom, I., Nicol, S. & de la Mare, W. (1996) Results of a hydroacoustic
survey of Antarctic krill populations in CCAMLR Division 58.4.1 carried out in January to
April 1996. SC-CAMLR WG-EMM-96/28.
Pauly, T., Nicol, S., Higginbottom, I., Hosie, G. & Kitchener, J. (2000) Distribution and
Distribution and Standing Stock
79
abundance of Antarctic krill (Euphausia Superba) off East Antarctic (80±1508E) during the
Austral summer of 1995/96. Deep-Sea Research II. (In press).
Priddle, J., Croxall, J.P., Everson, I., et al. (1988) Large scale fluctuations in distribution and
abundance of krill ± a discussion of possible causes. In Antarctic Ocean and Resources
Variability (D. Sahrhage, ed.), pp. 169±82. Springer-Verlag, Berlin.
Quetin, L.B., Ross, R.M., Frazer, T.K. & Haberman, K.L. (1996) Factors affecting distribution and abundance of zooplankton, with an emphasis on Antarctic Krill, Euphausia
superba. In Foundations for Ecological Research West of the Antarctic Peninsular, Vol. 70,
(R.M. Ross, E.E. Hofmann & L.B. Quetin, eds), pp. 357±71. American Geophysical Union,
Washington, D.C.
Ross, R.M. & Quetin, L.B. (1991) Ecological physiology of larval euphausiids, Euphausia
superba (Euphausiacea). Memoirs of the Queensland Museum 31, 321±33.
SC-CAMLR (2000) Report of the Nineteenth Meeting of the Scientific Committee.
CCAMLR, Hobart. (In preparation.)
Siegel, V. (1988) A concept of seasonal variation of krill (Euphausia superba) distribution and
abundance west of the Antarctic Peninsula. In Antarctic Ocean and Resources Variability
(D. Sahrhage, ed.), pp. 219±30. Springer-Verlag, Berlin.
Siegel, V., de la Mare, W. & Loeb, V. (1997) Long term monitoring of krill recruitment and
abundance indices in the Elephant Island area (Antarctic Peninsula). CCAMLR Sci. 4, 19±
35.
Siegel, V., Loeb, V. & Groger, J. (1998) Krill (Euphausia superba) density, proportional and
absolute recruitment and biomass in the Elephant Island region (Antarctic Peninsula)
during the period 1977 to 1997. Polar Biol. 19, 393±98.
Sushin, V.A. (1998) Distribution of the Soviet krill fishing fleet in the South Orkneys area
(Subarea 48.2) during 1989/1990. CCAMLR Sci. 5, 51±62.
Sushin, V.A. & Shulgovsky, K.E. (1999) Krill distribution in the western Atlantic sector of the
Southern Ocean during 1983/84 and 1987/88 based on the results of Soviet mesoscale
surveys conducted using an Isaacs-Kidd midwater trawl. CCAMLR Sci. 6, 59±70.
Trathan, P.N., Everson, I., Murphy, E.J. & Parkes, G.B. (1998) Analysis of haul by haul data
from the South Georgia krill fishery. CCAMLR Sci. 5, 9±30.
Trathan, P.N., Agnew, D.J., Miller, D.G.M., Watkins, J.L., Everson, I., Thorley, M.R.,
Murphy, E.J., Murray, A.W.A. & Goss, C. (1992) Krill biomass in Area 48 and Area 58:
recalculation of FIBEX data. In Selected Scientific Papers, 1992 (SC-CAMLR-SSP/9)
pp. 157±81. CCAMLR, Hobart, Australia.
Vagin, A.V., Makarov, R.R. & Menshinina, L.L. (1992) Dirunal variations in biological
characteristics of krill Euphausia superba Dana to the west of the South Orkney Islands 24
March to 18 June 1990 ± based on data reported by a biologist-observer. Scientific Committee for the Conservation of Antarctic Marine Living Resources, Selected Scientific Papers.
SC-CAMLR-SSP/9, pp. 201±222.
Watkins, J.L. (1999) A composite recruitment index to describe interannual changes in the
population structure of Antarctic krill at South Georgia. CCAMLR Sci. 6, 71±84.
Chapter 4
Aggregation and Vertical Migration
Jon Watkins
4.1
Introduction
Patchiness, or heterogeneity in the distribution of marine organisms is extremely
common across many spatial scales (Haury et al., 1978; Murphy et al., 1988; Mangel,
1994). In euphausiids patchiness is frequently manifested through the formation of
aggregations and according to Mauchline (1980) it is likely that all species of
euphausiids aggregate to some extent at certain times or locations. In species such as
Euphausia superba some of the most extreme examples of aggregation occur with a
substantial proportion of the population occurring in swarms where densities may
reach thousands of krill m±3. The formation of aggregations affects many aspects of
the euphausiid life cycle. Thus feeding, and hence growth, are affected by the
interaction between the patchy distributions of the krill and the food supply; similarly predation and hence mortality of euphausiids are influenced markedly by the
aggregation or diurnal migration of the krill. The phenomenon of aggregation in
euphausiids is also of great importance to commerical fisheries, for instance off
Tasmania if Nyctiphanes australis do not swarm then the jack mackerel do not
school in commercial quantities and that fishery fails (Harris et al., 1992). As a result
of these important interactions considerable work has been undertaken on the
phenomenon of euphausiid aggregation over the years which has been summarised
in a series of very detailed reviews; the first review discussing aggregation and
diurnal migration for all species of euphausiids was produced by Mauchline &
Fisher (1969) and this was then extended and updated by Mauchline (1980). In
addition to these works dealing with euphausiids in general there have been reviews
concentrating on E. superba (Miller and Hampton, 1989a; Siegel and Kalinowski,
1994). Most recently the Second International Krill Symposium held in Santa Cruz,
USA, in August 1999 held a session on krill swarming and aggregation.
Acoustic records show that krill aggregations vary from swarms or schools that
are dense and compact (only metres in any dimension) through to large layers,
aggregations or deep scattering layers that may be kilometres in horizontal extent.
The schools tend have a structure like fish schools with individuals showing parallel
orientation within the school while in swarms the orientation of individuals is
irregular (Ritz, 1994). Frequently the animals with the swarm or school are much
more similar to each other than they are to krill in neighbouring swarms, so animals
80
Aggregation and Vertical Migration
81
within a swarm may be of a very similar size or maturity stage or at the same stage in
the moult cycle (Watkins et al., 1992; Buchholz et al., 1996). The densest aggregations are likely to be formed and maintained through social interactions (Ritz,
1994). These krill aggregations are dynamic and this is reflected in the dispersal or
concentration of aggregations over various temporal scales. The aggregations are
capable of directed movements which in the vertical plane are often expressed as a
diurnal vertical migration, while large-scale horizontal movements have also been
suggested (Kanda et al., 1982). Previously diurnal migration and aggregation have
been considered quite separately (Mauchline & Fisher, 1969; Mauchline, 1980;
Miller & Hampton, 1989), indeed largely in different sections of these publications.
However, as Ritz (1994) notes it is quite likely that an ultimate benefit of both
swarming and diurnal vertical migration is protection from predation. Here,
therefore, they are considered together as differing behavioural strategies influenced by the same biotic and abiotic factors and having similar costs and benefits.
A major focus of this book is considering the processes of ecosystem assessment,
and here the focus is on aspects of aggregation and diurnal migration that are likely
to be relevant to that process. Therefore a major aim is to describe patterns of
aggregation and how these vary over different time and space scales. The comprehensive review on Euphausia superba by Miller & Hampton (1989a) contains
considerable information on aggregation in krill and, in addition, highlighted
priority issues for future work in the study of aggregations. Therefore how the study
of aggregation and vertical migration has progressed in the last ten years is considered, particularly with reference to key issues of describing aggregations, the
continuing problem of estimating density in aggregations and understanding the
spatial and temporal scales of aggregation, and areas where further work needs to be
carried out as a priority.
4.2
Physical and biological descriptions of krill aggregations:
classification or not?
Aggregation of euphausiids occurs over a range of temporal and spatial scales,
creating what is often considered to be a hierarchy of different kinds of aggregation,
from the dense school at the scale of metres to the patch or the concentration at the
scale of up to hundreds of kilometres (Murphy et al., 1988). Our ability to sample
these euphausiid aggregations over a range of temporal and spatial scales using a
variety of sampling devices, such as acoustics, direct visual observations and nets
(see Chapter 2), has led to various ways to summarise the vast amount of data.
Frequently this has been done by producing a classification scheme for different
types of aggregations (Mauchline, 1980; Kalinowski & Witek, 1985; Murphy et al.,
1988; Miller & Hampton, 1989a; Siegel & Kalinowski, 1994). The original scheme
used by Mauchline (1980) was developed for mysids and euphausiids, and utilised
physical parameters of size and density coupled with observations of internal
82
Krill: Biology, Ecology and Fisheries
structure, in particular the degree of parallel orientation between individuals within
the aggregation. With the increasing use of acoustic observation techniques this
scheme was modified by Kalinowski & Witek (1985) just to include parameters that
could be determined from an echo-chart produced by an echo-sounder. Their swarm
classification has been modified several times and the most recent manifestation is
that given by Siegel & Kalinowski (1994) and reproduced here in Fig. 4.1. The
scheme is hierarchical and typically there have been three levels of organisation
corresponding to characteristic spatial (and hence temporal) scales which were
originally defined by Haury et al. (1978). At the micro-scale (distances < 10 km)
aggregations consist of swarms, layers and scattered forms, irregular forms, nonaggregated forms and super-swarms. Each of these different types has a different
combination of size, shape and density (expressed as number of individuals or mass
m±3). Groups of swarms, layers or other micro-scale forms may be grouped together
at the meso-scale (distances 10±1000 km) to form patches and similarly groups of
patches may be grouped into a concentration of krill which covers spatial scales of
more than 1000 km.
In the case of E. superba, swarms are the most frequent type of aggregation
observed and the physical size of swarms from a number of studies has been summarised by Miller & Hampton (1989a, b), and Siegel & Kalinowski (1994). Typically
aggregations classified as swarms are small, for example the mean horizontal extent
of swarms recorded on nine summer cruises to the west of the Antarctic Peninsula
varied between 54 and 104 m and the mean vertical extent between 11 and 17 m
(Ross et al., 1996).
The acoustic record is very good at providing information on the physical
dimensions of krill swarms; however it cannot, as yet, provide any information on
the biological composition of the individual krill in the swarm or the behaviour of
the krill. Our information on these factors has come from two different kinds of
studies; high resolution net sampling and direct in situ observations. High resolution
net sampling in conjunction with acoustics has been used to obtain discrete samples
from within individual swarms that were in close proximity (Ricketts et al., 1992).
This leads to the conclusion that the krill within a swarm are more similar than krill
in adjacent swarms ± thus it might be said that proximity does not imply similarity
(Watkins, 1986; Watkins et al., 1986; Fig. 4.2). The relationship between the biological characteristics of krill in swarms, that is whether the krill are large or small,
moulting or not moulting, reproductively mature or immature, did not bear any
relationship to the acoustic description of the swarms (Ricketts et al., 1992).
According to Ritz (1994) this is not surprising given that social behaviour is likely to
be very important in determining what happens at the scale of individual krill
swarms (see also Hamner, 1984). For instance the description of the physical and
biological characteristics of krill swarms usually does not take account of the costs
and benefits to individuals that are in a swarm. The calculation of such costs and
benefits is not only complex but also very dynamic. In the case of searching for food,
groups appear to be more successful at finding patchily distributed food than do
t = 1–20m
1 = several 10's m
Swarms/schools
Cohesive
aggregations
t = 100–250m
1 = up to several km
Super–swarms
t = large
1 = many km
1.
t = 10's m
Irregular forms
Dispersed
aggregations
Layers and
scattered forms
2.
Non-aggregated
forms
p = <0.1 g . m –3
Non-aggregated
forms
2.
Micro
(< 10 km)
Fig. 4.1 Classification scheme for krill aggregations as modified by Siegel & Kalinowski (1994) from Kalinowski & Witek (1985) and Miller &
Hampton (1989a). t = thickness, l = intercepted length, p = density. Reproduced with permission from Cambridge University Press.
Hours to
days
1.
Meso
(10–1000 km)
Patches
Days to
months
2.
Macro
(> 1000 km)
Concentration
Months
1. = Shoats
2. = Other forms
Spatial Scale
Time Scale
Aggregation and Vertical Migration
83
57 00'W
61 10'S
56 40'W
56 20'W
56
48 49 50
4
29
30
21
61 20'S
27
5
25
28
23
22
24
17
1
2
9
500
(a)
11
12
Study
area
13
20
0
61 40'S
57
15
2
Gibbs Is.
61 30'
0
7
Elephant Is.
61 S
00
50
61 30'S
17 13 56 21
2
26
15
10
7
32
31
56
55 W
62
Hauls with 1 aggregation
1
4
5
9 10
12
32
23
31
22
24
49
25
–2
Hauls with 2 aggregation
11
26
48
Hauls with 3 aggregation
50
28
27 30 29
–1
0
Mean values of derived variable
Hauls with 6 aggregation
+1
Fig. 4.2 Similarities and differences between 28 discrete swarms (shown as numbered dots) of
Euphausia superba sampled during net hauls (shown as lines) near to Elephant Island, Antarctica.
The degree of similarity between swarms is summarised using a discriminant function variate to
combine variation due to interswarm differences in krill length, sex-ratio, maturity stage, alimentary tract fullness and exoskeleton hardness. Swarms that occur together in the same haul
(connected with a line) are frequently more different from each other than from swarms that occur
in different hauls. Modified with permission from Springer-Verlag, Watkins et al. (1986).
Aggregation and Vertical Migration
85
individuals (Pitcher et al., 1982; Grunbaum, 1998). Therefore it is advantageous to
swarm to find food but when food becomes limited and the swarm size is large the
returns per individual are probably reduced and so it might be better for an individual to leave the swarm. However, if the swarm is threatened by a predator the
best strategy may be for the swarm to remain as large as possible even if the food
returns per capita decline (Ritz, 1994). Such trade-offs between individual and
aggregation become even more complex in the case of some euphausiids where
swarming as a predator response does not provide equal protection against all
predators. Indeed in some cases predation, for instance by whales, can only take
place because krill swarm (Brodie et al., 1978; Murphy, 1995).
Information on the composition of krill swarms has also come from studies made
by divers; in this case there have been observations that the krill are of a similar size
but, more importantly has been the observation that in a swarm the individual krill
were all swimming in the same direction, i.e. the swarm is polarised or shows parallel
orientation (Hamner et al., 1983; Hamner, 1984; Stretch et al., 1988; Strand &
Hamner, 1990; O'Brien, 1987; Fig. 4.3). These observations of the biological and
behavioural properties of swarms of E. superba have lead to the suggestion at
various times that the krill swarm should be considered as the basic unit of organisation in krill ecology (Watkins, 1986; Murphy et al., 1988).
While the acoustic and direct observations seem to be to a large extent in
Fig. 4.3 Antarctic krill (Euphausia superba) in a swarm occurring on the northern shelf of South
Georgia. Photograph taken with Camera Alive remotely controlled underwater camera by author.
86
Krill: Biology, Ecology and Fisheries
agreement about the size and importance of krill swarms, the classification of other
types of aggregation is more problematic. Thus although acoustically we may see
features that are described as layers, irregular forms and superswarms (Kalinowski
& Witek, 1985), the distinctions between such forms are often very subjective. Far
less is known about the structural integrity of such features and it is much harder to
relate such large aggregations with the understanding of small krill swarms controlled by interactions between individuals. Recent analysis of the biological
properties and acoustic structure of some extensive krill layers now sheds some light
on these discrepancies. Thus Watkins & Murray (1998) report on the characteristics
of Euphausia superba in diffuse layers detected at night (see Fig. 4.4). These layers
have considerable physical structure with variations in depth and thickness of the
layer, indeed in some areas there appear to be minor gaps in the layer. The analysis
of the type of krill within the layer using a multi-serial sampler that sampled more
than 60 discrete samples shows that the variation in krill length, maturity and sexratio was as great as between discrete krill swarms that were detected within the
same area during the day time. This indicates that the layer could well be composed
of a series of krill swarms that had been brought so close together that they were
touching but that there had been little mixing. In this case it therefore still seems
valid to consider that these layers can be considered as a group of independent
swarms that maintain their integrity. Little information is available on the internal
biological structure of some of the other large-scale krill aggregations and it would
appear that no layers and superswarms have been observed by divers. Perhaps the
main reason for this is the relative rarity of such aggregations. Further information
on the structure of these layers is required and should be a priority for small-scale
and in situ observations to find out whether the heterogeneity within the layers
observed here is a common phenomenon.
As suggested above, the attempts to classify different types of aggregations can
lead to problems of interpretation. Indeed Miller & Hampton (1989a,b) point out
that the size range of Euphausia superba swarms is very large (see also Miller et al.,
1993) and this may be due to inclusion of some swarms that perhaps should have
been classified as layers. With the increasing availability of more computer intensive
analysis techniques it is suggested that we should worry less about swarm classification and rather put more time and effort into descriptive techniques that can take
account of large amounts of variation in spatial structure; examples that could prove
fruitful for future research include the fractal dimensions analysis undertaken by
Krause (1998) and also the increasing use of geostatistics (Murray, 1996).
Fig. 4.4 (Opposite.) The structure of a diffuse layer of Euphausia superba investigated using
acoustics and a multi-sample LLHPR (see Chapter 2). The number of krill taken in each sample
can be seen in relation to the depth of the net and the acoustic structure of the layer. The histograms show the variation in mean length of krill and percentage of male krill in each sample. Layer
1 is the layer shown at the top of the figure. Layer 2 was sampled on a separate night. The variability
in length and percentage of male krill within a single layer is as great as the variability occurring
between 28 discrete swarms sampled in the same area (see Fig. 4.2). Modified with permission from
Springer-Verlag, Watkins & Murray (1998).
370 m
Number of krill
400
50 m
200
0
10
20
30
40
Sample number
(a)
30
Percentage occurrence
25
Swarms
Layer 1
Layer 2
20
15
10
5
0
42
43
44
45
46
47
48
49
Mean length of krill in sample
30
(b)
25
Percentage occurrence
0m
Swarms
Layer 1
Layer 2
20
15
10
5
0
10
20
30
40
50
60
70
Percentage of male krill
80
90
100
88
Krill: Biology, Ecology and Fisheries
The behavioural observations of parallel orientation between individuals and
concerted predator avoidance behaviour (Hamner et al., 1983; Hamner, 1984; Ritz,
1994) indicate that individuals in these small swarms act in unison. If this is so, then
we might ask what is the maximum size of krill swarm that can exist and still be
considered as a single unit? Ritz (1994) discusses the idea that there is an optimal
group size for an aggregation but that the potential benefits to individuals wishing to
join the group may push the size of the aggregation beyond the optimal size. If this is
the case then can the krill control their group size (Levin, 1994)? Individual krill
may be aware of the density of surrounding krill from information on nearest
neighbour distance but beyond some critical size of swarm may have no control over
the total group size (Levin, 1994). How does this critical size relate to the size of
small swarms which are seen so frequently and to what extent do such basic swarm
units then interact in larger aggregations? There seems to be a good case for
bringing the fields of behavioural observations and remotely observed phenomenon
together to increase our understanding of the way in which swarming works.
4.3
Aggregation density: differences and extremes
True densities in swarms
One of the key physical parameters used to describe krill aggregations is the density (usually expressed as number or mass of individual m±3). In Table 4.1 a compilation of the estimates of density of krill swarms using acoustic and visual
Table 4.1 Estimates of krill density in swarms.
Species
Density
Collection method
References
Euphausia superba
20 000±60 000 n m±3
scuba observations
Ragulin, 1969,
Hamner et al., 1983
Euphausia vallentini
70 000±80 000 n m±3
scuba observations
Ragulin, 1969
Meganyctiphanes
norvegica
9000±770 000 n m
±3
visual observations
and bag sampler
Nicol, 1986
Nyctiphanes australis
3000±480 000 n m±3
visual observations
O'Brien, 1988
visual observations
Hanamura et al., 1984
visual observations
Nicol et al., 1987
±3
Euphausia pacifica
10±72 000 n m
Euphausia lucens
4000±5 000 000 n m±3
±3
Euphausia superba
150±500 g m
(latest TS used)
acoustic survey
Lascara et al., 1999
Euphausia superba
300±800 g m±3
(BIOMASS TS)
acoustic survey
Macaulay et al., 1984
Euphausia superba
1534 g m±3
(BIOMASS TS)
acoustic survey
Higginbottom &
Hosie, 1989
Aggregation and Vertical Migration
89
techniques has been produced; net haul estimates have been excluded from this
table because of the effects of net avoidance (see Chapter 2). Visual estimates of
Euphausia superba swarms may be up to 20 000 krill (Hamner et al., 1983) or
even as high as 60 000 krill m±3 while the highest published acoustic densities for
Antarctic krill are in the region of 1500 krill m±3 (Higginbottom & Hosie, 1989).
Densities from Ritz (1994), summarising visual and photographic density estimates for other species of euphausiid are even higher (Table 4.1), for instance up
to 480 000 krill m±3 for Nyctiphanes australis and as high as 5 000 000 krill m±3 for
Euphausia lucens.
While there is considerable variety in density estimates, however, it is generally
acknowledged (Miller & Hampton, 1989a; Siegel & Kalinowski, 1994) that visual
or photographically based estimates are considerably higher than acoustic estimates. From the data presented in Table 4.1 this difference may be as much as
two orders of magnitude. Such differences are likely to be due to a combination
of behavioural and methodological differences. Some of these differences may be
explained by the considerable differences in the degree of dispersion that has
been observed over a diurnal cycle (see for instance Everson, 1982). However
some of these differences are undoubtably due to methodological effects. Accurate acoustic density estimates depend upon precise calibration and a realistic
Target Strength (TS) relationship (see Chapter 2 for further details). The acoustic
techniques have evolved considerably over the two decades in which observations
on Antarctic krill have been collected and the present day acoustic data are generally exhaustively calibrated and validated (Chapter 2, Watkins & Murray, 1998;
Foote et al., 1987). However, the re-evaluation of TS in the early 1990s has had a
major change on the calculation of krill biomass (Everson et al., 1990; Foote et
al., 1990), thus all of the acoustic swarm density estimates quoted by Miller &
Hampton (1989a, b) and many of those quoted by Siegel & Kalinowski (1994)
use the TS derived during the BIOMASS Programme. This TS relationship
underestimates the density by as much as an order of magnitude (Everson et al.,
1990; Trathan et al., 1995).
Recent estimates of swarm density have been calculated using the most up-todate TS relationships (Lascara et al., 1999) but these density estimates are still low in
comparison to the diver observed densities (Table 4.1). However, note that the
highest published density estimate of the swarm observed acoustically by Higginbottom & Hosie (1989) was also calculated using the BIOMASS TS relationship;
assuming that the more recent TS would increase the density fivefold results in a
density estimate in the region of 9000 krill m±3 (also assuming that the weight of an
individual krill is 1 g). Such a value is much closer to the visual density estimates.
The accuracy of the diver density estimates is harder to determine; Ritz (1994) states
that `visual estimates and those made from plankton net catches are notoriously
unreliable'. Visual and photographic density estimates are usually calculated using a
packing model which requires the measurement of nearest neighbour distance
(NND ± the distance between adjacent krill). There is little information on precision
90
Krill: Biology, Ecology and Fisheries
or bias of estimates of NND (Watkins & Murray, 1998), although O'Brien (1989)
indicates that some packing models may overestimate density by 20%. It is maintained therefore that there is still considerable work to be done to achieve a consensus on the range of density estimates that are found in krill swarms and
aggregations, and studies using several independent methods of estimating krill
swarm densities should be carried out.
The importance of `huge' aggregations
The extensive acoustic data sets collected over the last 30 years (see for instance
Miller & Hampton, 1989a, b; Miller et al., 1993; Ross et al., 1996; Lascara et al., 1999)
have sampled many thousands of krill aggregations and confirm, at least for
Euphausia superba, that the majority of krill aggregations occur as small, discrete
swarms (Miller & Hampton, 1989a, b; Watkins, 1986). The larger aggregations of
Euphausia superba which occur as layers or irregular forms tend to be structures
with a density substantially lower than the small swarms (Kalinowski & Witek, 1985;
Watkins & Murray, 1998). However, now there does seem to be accumulating
evidence that occasionally krill are found in aggregations that are both very
extensive and dense. Such huge aggregations would appear to be relatively rare, for
instance Higginbottom & Hosie (1989) carried out an acoustic survey of the Prydz
Bay region in Antarctica which covered 1.3 million km and describe the occurrence
of a large, dense aggregation with an estimated area of 5.7 km2 (Table 4.2). This
aggregation was two orders of magnitude more dense than any other aggregation
observed in a total of six previous cruises carried out by the Australian Antarctic
Division. Another dense and even more extensive superswarm of Euphausia
superba off Elephant Island has been described in considerable detail (Mathisen &
Table 4.2 Characteristics of large aggregations of Euphausia superba.
Horizontal
area
(km2)
Vertical
extent
(m)
Average
(max) density
(g m±3)
Total
biomass
(tonnes)
Higginbottom Prydz Bay
& Hosie, 1989
5.7
60±80
94.7 (1534)*
57 000*
Macaulay et
al., 1984
Elephant Island
150
up to 250 m
300±500
(up to 800)*
> 1.2* 106*
Murray et al.,
1995
Bellingshausen
Sea
NA
70 (±)
37 000
Watkins,
Murray &
Brierley,
unpublished
South Georgia
2.8 (distance
not area)
0.9 (distance
not area)
50
230 (±)
145 000
Reference
Location
* Note these calculations are based on the BIOMASS Target Strength relationship, use of the more recent
CCAMLR TS relationship would increase the values by a factor of between 5 and 10.
Aggregation and Vertical Migration
91
Macaulay, 1983; Macaulay et al., 1984) (Table 4.2). The occurrence of such large
aggregations does not appear to be restricted only to Euphausia superba, thus
Simard & Lavoie (1999) describe a large aggregation of Meganyctiphanes norvegica
and Thysanoessa rachi at the head of the Gulf of St. Lawrence.
These aggregations contain substantial quantities of krill. Indeed the single dense
swarm observed by Watkins, Murray & Brierley (BAS unpublished data) in 1997
around South Georgia accounted for more than 12% of the total biomass in an area
of 80 000 km2. Analyses of acoustic survey data in the shelf waters of the Antarctic
Peninsula, showed that more than 80% of the krill biomass was associated with a
small number of aggregations (less than 20% of the total) that were two- to four-fold
larger and an order of magnitude more dense than average (Lascara et al., 1999).
The occurrence of such aggregations raises a number of important issues with regard
to survey design and analysis, aggregation formation and implications for the
functioning of the ecosystem.
In terms of ecosystem assessment, such aggregations create significant problems
for the design of surveys to calculate biomass. The area occupied by such swarms is
still small in relation to the spacing between transects frequently used by acoustic
surveys, which may vary from about 10 km (Brierley et al., 1997) to more than 100 km
(Lascara et al., 1999; Pauly et al., 1996). Therefore the probability of encountering
such aggregations would appear to be quite small. When encountered within a survey
such large concentrations of biomass also cause significant problems for the
assessment of variance of the biomass estimate and probably need to be treated
separately from the usual transect data (Macaulay et al., 1984; Murray, 1996).
How do such large aggregations arise? Because of their relative rarity we know
little about the conditions under which such swarms form. Is this simply through the
amalgamation of smaller swarms as was suggested for the low density layer
described by Watkins & Murray (1998) or is it necessary for physical concentration
of swarms into an area to occur as suggested by Witek et al. (1988). Interactions with
the physical environment may be important in the formation of at least three of the
huge aggregations described here. The Elephant Island superswarm was found on
the shelf break (Macaulay et al., 1984) while the Prydz Bay swarm was associated
with a small seamount (Higginbottom & Hosie, 1989). The aggregation of
euphausiids along the northern edge of the Laurentian Channel in the St. Lawrence
Estuary may have resulted from the interaction between the negative phototaxis of
the euphausiids and upwelling of the water circulation in the area (Simard et al.,
1986). Social interaction between krill is very important in the formation and
maintenance of small swarms (Ritz, 1994) but both the role in these larger aggregations together with the possible role of predators on the formation of such large
aggregations need to assessed. The dense swarm off South Georgia described above
was seen in association with southern right whales and many fur seals (E.J. Murphy,
pers. comm.). It may have been that the large, high density swarm was produced as a
result of predator avoidance behaviour by the krill or herding behaviour by the
whales (Hamner et al., 1988).
92
Krill: Biology, Ecology and Fisheries
In addition such aggregations are likely to have a great impact on predator prey
interactions. Thus for instance in the Bransfield Strait, Antarctica, during a 20-day
cruise over 60% of all the Cape Petrels and Antarctic Fulmars were seen at one krill
aggregation and more than 75% of all Adelie Penguins were seen at another large
krill aggregation (Heinemann et al., 1989; Hunt, 1991). However, the duration and
subsequent fate of such aggregations is not known. Therefore although such swarms
may be very important when they occur, if such swarms only occur for a few days at
a time and are relatively infrequent then their importance to predators over a
breeding cycle may be far less important.
Given the importance to our understanding of krill biology, in particular the
relative roles of physical and biological forces in the formation and maintenance of
such aggregations, and also for assessing the status of the ecosystem it is vital that
studies on these large aggregations should be undertaken whenever they are
encountered as a matter of high priority.
4.4
Temporal and spatial structure
Diurnal time-scales
Much has already been written on the diurnal variation in the behaviour of krill
aggregations (Mauchline & Fisher, 1969; Mauchline, 1980; Siegel & Kalinowski,
1994; Everson, 1982; Everson, 1983). More than 75% of all euphausiid species
undertake some form of diurnal vertical migration (Mauchline, 1980) and this has
been extensively documented for species such as Meganyctiphanes norvegica (see
Onsrud & Kaartvedt, 1998; Buchholz et al., 1995; Tarling et al., 1998). In Euphausia
superba krill commonly show a diurnal vertical migration towards the surface at night
which may be linked to dispersal of the swarms at this time (see also Demer & Hewitt,
1995; Ross et al., 1996; Godlewska, 1996). However, there can be considerable
variation from this basic pattern. For instance krill swarms may remain as distinct
swarms throughout the day and night (Watkins et al., 1992) or swarms may be found at
the surface during the daytime (Marr, 1962). It would appear that all krill that swarm
also vertically migrate at least to some extent (Mauchline, 1980), and it is likely that
both diurnal migration and swarming are linked to feeding or predator defence
mechanisms (Ritz, 1994). It is therefore interesting to consider why these species
exhibit both strategies; Ritz (1994) speculates that social (i.e. swarming) euphausiids
might show different vertical responses to non-social ones and suggests that diurnal
vertical migration in the facultative schoolers might be different when dispersed
compared to while schooling all the time. However, it is not clear that this is occurring
in the case of the seasonal differences in swarming and diurnal vertical migration
presented by Ross et al. (1996) (see next section). It is likely that the complex and
variable patterns of behaviour shown by swarms of Euphausia superba are an
interplay between the various advantages and disadvantages resulting from aggre-
Aggregation and Vertical Migration
93
gation and vertical migration, and that such balances change under differing
environmental conditions and predation pressure. To date there has been no attempt
to produce an integrated synthesis of diurnal vertical migration and swarming; such
an undertaking, although daunting, is to be encouraged.
Godlewska (1996) has produced a very comprehensive synthesis of diurnal vertical migration in Euphausia superba using both data collected during Polish Antarctic cruises and also analysing much of the published literature. Vertical migration
has been described by a model which combines the amplitudes of migration
detected with a 12- and 24-hour periodicity. This analysis shows that the amplitude
of vertical migration in Euphausia superba varies from only a few to 30 m. Such
migrations are very small in comparison with distances of 400+ m described for
tropical euphausiids (Andersen et al., 1997). Food availability appears to be one of
the most important factors affecting the extent of the migration amplitude; Godlewska (1996) found that under good feeding conditions the amplitude is large and
duration is dominated by a 24-hour periodicity. As food becomes scarce, the
amplitude decreases and the migration cycle has a greater component of the 12-hour
periodicity (Godlewska, 1996). Once again, however, the complex interactions
between the underlying causes and expression of diurnal migration are illustrated by
the observation that the amplitude of vertical migration in euphausiid species at
meso- and eutrophic stations in the north-east tropical Atlantic was significantly less
than that observed in a low production oligotrophic station (Andersen et al., 1997).
The role of light in initiating diurnal vertical migration has long been recognised (see Ringelberg, 1995). Godlewska (1996) confirms that in Euphausia
superba, changes in light intensity rather than the actual light level appear to trigger the migration process. Other biotic and abiotic factors then serve to modify
the extent of the migration. Such a process also appears to control the migrations
occurring in other species. Thus Onsrud & Kaartvedt (1998) discuss the role of
the physical environment, food and predators in modifying the diurnal vertical
migration of Meganyctiphanes norvegica. Their results indicated that M. norvegica migrated irrespective of the level of phytoplankton or the presence of copepod prey at the daytime depths but that night-time migration into the surface
layer was restricted when planktivorous fish were most abundant at the surface.
Interestingly Onsrud & Kaartvedt (1998) also note that the migrating scattering
layer appeared to follow a characteristic isolume during the ascent towards the
surface which would appear to be unusual (cf. Ringelberg, 1995; Roe, 1983). The
modifying effects of the physical environment on vertical migration may have
profound implications for the distribution of euphausiids. In Gullsmarsfjorden
(Sweden) M. norvegica did not migrate through a pycnocline at 50±60 m while
Thysanoessa raschii did migrate up into the water above the sill (Bergstrom &
Stromberg, 1997). Such differences in behaviour may have a marked effect on the
horizontal distribution of species (see Kaartvedt, 1993). In addition to direct reactions to light level or change in light intensity, it is possible that migration is cued
by an endogenous rhythm. The most recent evidence pointing to this comes from
94
Krill: Biology, Ecology and Fisheries
investigation of the effect of a lunar eclipse on the diurnal vertical migration in
M. norvegica (Tarling et al., 1999). In this study the secondary midnight sinking
frequently observed in euphausiids was linked to the time of moon rise, but did
not occur on the evening of a lunar eclipse when the moon effectively did not
rise. Behaviour later in the month, when midnight sinking occurred at a relatively
constant time and prior to moon rise, was attributed to an endogenous rhythm
which was synchronised by the external cue.
Most of our information on diurnal migration and swarming is collected by
observing a number of swarms for short periods of time; in the case of acoustic
observations it is usually the time taken to steam over an individual aggregation,
while in the case of diver observations the swarm is only likely to be observed for a
matter of minutes. Consequently our knowledge of swarm dynamics is built up from
a series of snapshots of many different swarms rather than an understanding of how
individual swarms behave over longer periods of time. Many of the questions about
how krill swarms form, maintain themselves and interact with the environment at
small scales will probably only be resolved once extended observations on single
swarms are carried out. The increased availability of non-invasive observation
techniques (see Chapter 2 for more details) is likely to lead to substantial advances
in this field. For instance, the use of ADCPs to measure the actual migration rates of
individual scattering layers or swarms is becoming increasingly common (see for
instance Tarling et al., 1998; Tarling et al., 1999).
Seasonal variation in aggregation and diurnal vertical migration
Most studies of aggregation and diurnal migration have been one-off studies and
any information about seasonal changes has come from the comparison of cruises
carried out to different protocols. Recently however, two major studies have been
published where exactly the same methods and protocols have been used over a
number of seasons (Ross et al., 1996; Lascara et al., 1999). Ross et al. (1996) carried
out 12 cruises over the same acoustic survey grid west of the Antarctic Peninsula
around Anvers Island and within the Bransfield Strait from February 1985 to
December 1988. They observed aggregations of Euphausia superba throughout the
year but there were seasonal differences in the number of aggregations, the total
cross-sectional area of aggregations per kilometre and the size of the aggregations.
During the spring and summer an increase in the total cross-sectional area per unit
distance was attributed to the increase in the number of swarms. Maximum total
cross-sectional area per unit distance occurred in the winter when the actual number
of swarms seen was low but the size of the aggregations was very high (Fig. 4.5). The
results of spring, summer, autumn and winter cruises using another survey grid west
of the Antarctic Peninsula between 1991 and 1993 (Lascara et al., 1999) revealed a
cyclical change in aggregation dimensions and density similar to those observed by
Ross et al. (1996). In the spring aggregations typically had cross-sectional areas
between 700 and 2200 m2 and mean biomass densities between 20 and 100 g m±3; the
Cross-sectional area per distance
(m 2 km –1)
Aggregation and Vertical Migration
(a)
600
500
400
300
200
100
0
25
Schools per distance travelled
(No. 100 km –1)
95
(b)
20
15
10
5
0
Median cross-section area (m 2)
(c)
12 000
8000
4000
0
Nov
Jan
Mar
May
July
Sept
Month
Fig. 4.5 Seasonal variation in Antarctic krill aggregation parameters collected on 12 Antarctic
cruises to the west of the Antarctic Peninsula. Aggregation statistics for each cruise have been
normalised to the total distance travelled. (a) vertical cross-sectional area (m2) detected per km; (b)
number of aggregations detected per km; (c) median cross-sectional area of all aggregations within
a cruise. Figure from Ross et al. (1996).
96
Krill: Biology, Ecology and Fisheries
smallest densest aggregations were found in the summer (area < 1000 m2; mean
biomass density 150±500 g m±3), and then aggregations became larger and less dense
through the autumn and by winter were characterised by aggregations with crosssectional areas > 7000 m2 and mean biomass densities < 10 g m±3.
In these two studies there were seasonal differences in the depth distribution of
swarms but the pattern observed was slightly different. Lascara et al. (1999) detected
a seasonal shift in the depth distribution of the krill swarms. In the summer krill
swarms were concentrated in the top 70 m of the water column while in the winter a
substantial proportion of the swarms were found below 100 m. In the spring (and to
a lesser extent during the autumn) swarms were still found in the top 70 m but the
range of depths over which the swarms were found was greater. In contrast Ross et
al. (1996) found that the overall depth distribution of the swarms did not change
throughout the year; however, the pattern of diurnal vertical migration changed,
with krill showing a marked diurnal migration only during the spring and autumn
(Fig. 4.6). Seasonal patterns in diurnal vertical migration of other euphausiids have
also been described. For instance in the Kattegat the timing of night-time dispersal
and migration of Meganyctiphanes norvegica varied with day length, the period of
dispersion being longer in the winter (Tarling et al., 1998).
Spatial structure
For many observers krill swarms are likely to be seen only in two dimensions. The
acoustic techniques in common use produce a vertical two-dimensional slice through
the swarms (see for instance Everson, 1982; Everson, 1987). Similarly the observation
of surface swarms produces a two-dimensional plan view of the shape (Nicol, 1986).
Surface swarms are frequently very irregular in shape (Marr, 1962) and so any
assumptions about the shape of a swarm in the unobserved third dimension may be
very misleading. This is supported by the relatively limited number of observations of
swarms detected by divers which reveal that frequently even the large and dense krill
swarms are narrow in one dimension (Hamner, 1984). While knowledge of the shape
of swarms may not be needed for biomass estimates using classical sampling
methodology, the ability to build up a true three-dimensional picture of krill swarms
would be immensely valuable to increase our understanding of small-scale spatial
structure of krill swarms. Multi-beam sonar systems are now being used to make 3D
observations on fish schools (Misund et al., 1995; Gerlotto et al., 1999) and may
provide a suitable way of observing krill swarms in the near future.
4.5
Important processes in inaccessible regions
The frequent utilisation of acoustic technology has resulted in an avalanche of
detailed information on aggregation characteristics over large areas of the ocean.
However, there are a number of areas that are relatively inaccessible to the com-
Aggregation and Vertical Migration
97
Dec 1985
0
60
120
Feb 1985
0
60
120
Mar 1985
0
60
School depth (m)
120
Mar/April 1985
0
60
120
Jun/Jul 1987
0
60
120
0000
0400
0800
1200
1600
2000
2400
Time (GMT)
Fig. 4.6 Seasonal variation in the diurnal vertical distribution of average depths of swarms of
Euphausia superba from 5 cruises west of the Antarctic Peninsula. The shaded area is from sunset
to sunrise for 648S. Figure from Ross et al. (1996).
monly used acoustic instruments. Two that are of considerable importance for
understanding interactions within the ecosystem are the under-ice environment
inhabited by Euphausia superba and the sea surface which may be inhabited by
many euphausiid species during regular excursions to the surface at night or by
euhausiids swarming at the surface during the day (Komaki, 1967; Nicol, 1984).
98
Krill: Biology, Ecology and Fisheries
It is immediately obvious why the under-ice environment is difficult to sample;
observations made by divers (O'Brien, 1987) and by ROVs (Bergstrom et al.,
1990) are time-consuming and generally limited in extent. Penetration by ships is
costly, time-consuming and disruptive of the very environment to be studied.
However, a significant proportion of the population of Antarctic krill spend part
of their life cycle under the ice (Siegel & Loeb, 1995) and so there is a requirement to understand how krill swarms interact with the ice environment. The limited information available to date suggests that there may be quite different
behavioural strategies. Thus krill may be dispersed and found in close association
with the sea-ice (Marschall, 1988) while at other times krill are found in dense
swarms well below the bottom of the ice (Daly & Macaulay, 1991). Once again it
is likely that the actual behaviour patterns observed are due to the dynamic
balance between predation risks and feeding requirements. The degree of interaction of krill with the ice environment is likely to have significant effects on
both the survival and large-scale distribution of krill (Murphy et al., 1998). It is
therefore very important that further investigations on krill in relation to ice
should be carried out. It may be less obvious why the surface layer should pose
such a sampling problem, especially when we consider that much of the original
information on swarming came from the observations of surface swarms (Hardy,
1936; Marr, 1962). In this case, however, the fact that transducers of acoustic
systems are traditionally mounted in the bottom of the ship's hull (probably at
least 5 m below the surface) means that often the only way to observe surface
swarms is visually. This again is time-consuming and frequently only carried out
in specialised studies. Daytime surface swarming has been described in a variety
of species and it has been attributed to a number of causes related to feeding,
reproduction and predator avoidance (Komaki, 1967). Such swarming not only
affects regular biomass estimates made as part of ecosystem assessments, it also
has a significant effect on the interactions between krill and predators.
Future research on krill swarming needs to address the processes occurring in
both these relatively inaccessible regions. One possible area for development is the
use of Autonomous Underwater Vehicles (AUVs) which may be used to collect
information on swarms over larger spatial scales in a non-disruptive manner (Fernandes et al., 2000).
References
Andersen, V., Sardou, J. & Gasser, B. (1997) Macroplankton and micronekton in the
northeast tropical Atlantic: abundance, community composition and vertical distribution in
relation to different trophic environments. Deep-Sea Res. I 44, 193±222.
Bergstrom, B. & Stromberg, J.O. (1997) Behavioural differences in relation to pycnoclines
during vertical migration of the euphausiids Meganyctiphanes norvegica (M. Sars) and
Thysanoessa raschii (M. Sars). J. Plankton Res. 19, 255±61.
Bergstrom, B.O., Hempel, G., Marschall, H.-P., North, A.W., Siegel, V. & Stromberg, J.-O.
Aggregation and Vertical Migration
99
(1990) Spring distribution, size composition and behaviour of krill Euphausia superba in
the western Weddell Sea. Polar Rec. 26, 85±89.
Brierley, A.S., Watkins, J.L. & Murray, A.W.A. (1997) Interannual variability in krill
abundance at South Georgia. Mar. Ecol. Prog. Ser. 150, 87±98.
Brodie, P.F., Sameoto, D.D. & Sheldon, R.W. (1978) Population densities of euphausiids off
Nova Scotia as indicated by net samples, whale stomach contents, and sonar. Limnol.
Oceanogr. 23, 1264±67.
Buchholz, F., Buchholz, C., Reppin, J. & Fischer, J. (1995) Diel vertical migrations of
Meganyctiphanes norvegica in the Kattegat ± comparison of net catches and measurements
with acoustic Doppler current profilers. Helgolander Meeresuntersuchungen 49, 849±66.
Buchholz, F., Watkins, J.L., Priddle, J., Morris, D.J. & Ricketts, C. (1996) Moult in relation to
some aspects of reproduction and growth in swarms of Antarctic krill, Euphausia superba.
Mar. Biol. 127, 201±208.
Daly, K.L. & Macaulay, M.C. (1991) Influence of physical and biological mesoscale dynamics
on the seasonal distribution and behavior of Euphausia superba in the Antarctic marginal
ice zone. Mar. Ecol. Prog. Ser. 79, 37±66.
Demer, D.A. & Hewitt, R.P. (1995) Bias in acoustic biomass estimates of Euphausia superba
due to diel vertical migration. Deep-Sea Res. I 42, 455±75.
Everson, I. (1982) Diurnal variation in mean volume backscattering strength of an Antarctic
krill (Euphausia superba) patch. J. Plankton Res. 4, 155±61.
Everson, I. (1983) Variations in vertical distribution and density of krill swarms in the vicinity
of South Georgia. Mem. Natl Inst. Polar Res., Spec. Issue 27, 84±92.
Everson, I. (1987) Some aspects of the small scale distribution of Euphausia crystallorophias.
Polar Biol. 8, 9±15.
Everson, I., Watkins, J.L., Bone, D.G. & Foote, K.G. (1990) Implications of a new acoustic
target strength for abundance estimates of Antarctic krill. Nature 345, 338±40.
Fernandes, P.G., Brierley, A.S., Simmonds, E.J., Millard, N.W., McPhail, S.D., Armstrong, F.,
Stevenson, P. & Squires, M. (2000) Fish do not avoid survey vessels. Nature 404, 35±36.
Foote, K.G., Everson, I., Watkins, J.L. & Bone, D.G. (1990) Target strengths of Antarctic
krill (Euphausia superba) at 38 and 120 kHz. J. Acoust. Soc. Am. 87, 16±24.
Foote, K.G., Knudsen, H.P., Vestnes, G., MacLennan, D.N. & Simmonds, E.J. (1987) Calibration of acoustic instruments for fish density estimation: A practical guide. ICES
Cooperative Research Report 144, 1±69.
Gerlotto, F., Soria, M. & Freon, P. (1999) From two dimensions to three: the use of multibeam sonar for a new approach in fisheries acoustics. Can J. Fish. Aquat. Sci. 56, 6±12.
Godlewska, M. (1996) Vertical migrations of krill (Euphausia superba Dana). Pol. Arch.
Hydrobiol. 43, 9±63.
Grunbaum, D. (1998) Schooling as a strategy for taxis in a noisy environment. Evol. Ecol. 12,
503±522.
Hamner, W.M. (1984) Aspects of schooling of Euphausia superba. J. Crustacean Biol. 4, 67±74.
Hamner, W.M., Hamner, P.P., Strand, S.W. & Gilmer, R.W. (1983) Behavior of Antarctic
krill, Euphausia superba: chemoreception, feeding, schooling, and molting. Science 220,
433±35.
Hamner, W.M., Stone, G.S. & Obst, B.S. (1988) Behavior of southern right whales,
Eubalaena australis, feeding on the Antarctic krill, Euphausia superba. Fish. Bull. 86,
143±50.
100
Krill: Biology, Ecology and Fisheries
Hanamura, Y., Endo, Y. & Taniguchi, A. (1984) Underwater observations on the surface
swarm of a euphausiid, Euphausia pacifica in Sendai Bay, northeastern Japan. La mer 22,
63±68.
Hardy, A.C. (1936) Observations on the uneven distribution of oceanic plankton. Discovery
Rep. 9, 511±38.
Harris, G.P., Griffiths, F.B. & Clementson, L.A. (1992) Climate and the fisheries off Tasmania ± interactions of physics, food-chains and fish. S. Afr. J. mar. Sci. 12, 585±97.
Haury, L.R., McGowan, J.A. & Wiebe, P.H. (1978) Patterns and processes in the time-space
scales of plankton distributions. In Spatial Patterns in Plankton Communities (J. Steele,
ed.), pp. 277±327. Plenum Press, New York.
Heinemann, D., Hunt, G. & Everson, I. (1989) Relationships between the distributions of
marine avian predators and their prey, Euphausia superba, in Bransfield Strait and
southern Drake Passage, Antarctica. Mar. Ecol. Prog. Ser. 58, 3±16.
Higginbottom, I.R. & Hosie, G.W. (1989) Biomass and population structure of a large
aggregation of krill near Prydz Bay, Antarctica. Mar. Ecol. Prog. Ser. 58, 197±203.
Hunt, G.L. (1991) Marine ecology of seabirds in polar oceans. Am. Zoologist 31, 131±42.
Kaartvedt, S. (1993) Drifting and resident plankton. Bull. mar. Sci. 53, 154±59.
Kalinowski, J. & Witek, Z. (1985) BIOMASS Handbook, No. 27: Scheme for classifying
aggregations of Antarctic krill. SCAR, Cambridge.
Kanda, K., Takagi, K. & Seki, Y. (1982) Movement of the larger swarms of Antarctic krill
Euphausia superba off Enderby Land during 1976±1977 season. J. Tokyo Univ. Fish. 68, 25±
42.
Komaki, Y. (1967) On the surface swarming of Euphausiid crustaceans. Pacific Sci. 21, 433±48.
Krause, D.C. (1998) Implications of a fractal distribution of plankton patchiness. In IOC
Workshop Report No. 142. pp. 220±32. UNESCO, Paris.
Lascara, C.M., Hofmann, E.E., Ross, R.M. & Quetin, L.B. (1999) Seasonal variability in the
distribution of Antarctic krill, Euphausia superba, west of the Antarctic Peninsula. DeepSea Research Part I ± Oceanographic Research Papers 46, 951±84.
Levin, S.A. (1994) Patchiness in marine and terrestrial systems: from individuals to populations. Phil. Trans. R. Soc. Lond. Ser. B 343, 99±103.
Macaulay, M.C., English, T.S. & Mathisen, O.A. (1984) Acoustic characterisation of swarms
of Antarctic krill (Euphausia superba) from Elephant Island and Bransfield Strait. J.
Crustacean Biol. 4 (Spec. No. 1), 16±44.
Mangel, M. (1994) Spatial patterning in resource exploitation and conservation. Phil. Trans.
R. Soc. Lond. Ser. B 343, 93±98.
Marr, J.W.S. (1962) The natural history and geography of the Antarctic krill (Euphausia
superba Dana). Discovery Rep. 32, 33±464.
Marschall, H.-P. (1988) The overwintering strategy of Antarctic krill under the pack-ice of the
Weddell Sea. Polar Biol. 9, 129±35.
Mathisen, O.A. & Macaulay, M.C. (1983) The morphological features of a super swarm of
krill, Euphausia superba. Mem. Natl. Inst. Polar Res., Spec. Issue 27, 153±64.
Mauchline, J. (1980) The biology of mysids and euphausiids. Adv. mar. Biol. 18, 1±681.
Mauchline, J. & Fisher, L.R. (1969) The biology of euphausiids. Adv. mar. Biol. 7, 1±454.
Miller, D.G.M. & Hampton, I. (1989a) BIOMASS Scientific Series, No. 9: Biology and
Ecology of the Antarctic krill (Euphausia superba Dana): A Review. SCAR and SCOR,
Scott Polar Research Institute, Cambridge, England.
Aggregation and Vertical Migration
101
Miller, D.G.M. & Hampton, I. (1989b) Krill aggregation characteristics: spatial distribution
patterns from hydroacoustic observations. Polar Biol. 10, 125±34.
Miller, D.G.M., Barange, M., Klindt, H., Murray, A.W.A., Hampton, I. & Siegel, V. (1993)
Antarctic krill aggregation characteristics from acoustic observations in the South West
Atlantic Ocean. Mar. Biol. 117, 171±83.
Misund, O.A., Aglen, A. & Fronaes, E. (1995) Mapping the shape, size and density of fish
schools by echo integration and a high resolution sonar. ICES J. mar. Sci. 52, 11±20.
Murphy, E.J. (1995) Spatial structure of the Southern Ocean ecosystem: predator-prey
linkages in Southern Ocean food webs. J. Anim. Ecol. 64, 333±47.
Murphy, E.J., Morris, D.J., Watkins, J.L. & Priddle, J. (1988) Scales of interaction between
Antarctic krill and the environment. In Antarctic Ocean and Resources Variability (D.
Sahrhage, ed.), pp. 120±130. Springer-Verlag, Berlin.
Murphy, E.J., Watkins, J.L., Reid, K., Trathan, P.N., Everson, I., Croxall, J.P., Priddle, J.,
Brandon, M.A., Brierley, A.S. & Hofmann, E. (1998) Interannual variability of the South
Georgia marine ecosystem: biological and physical sources of variation in the abundance of
krill. Fish. Oceanogr. 7, 381±90.
Murray, A.W.A. (1996) Comparison of geostatistical and random sample survey analyses of
Antarctic krill acoustic data. ICES J. mar. Sci. 53, 415±21.
Nicol, S. (1984) Population structure of daytime surface swarms of the euphausiid
Meganyctiphanes norvegica in the Bay of Fundy. Mar. Ecol. Prog. Ser. 18, 241±51.
Nicol, S. (1986) Shape, size and density of daytime surface swarms of the euphausiid
Meganyctiphanes norvegica in the Bay of Fundy. J. Plankton Res. 8, 29±39.
Nicol, S., James, A. & Pitcher, G. (1987) A first record of daytime surface swarming by
Euphausia lucens in the southern Benguela region. Mar. Biol. 94, 7±10.
O'Brien, D.P. (1987) Direct observations of the behaviour of Euphausia superba and
Euphausia crystallorophias (Crustacea: Euphausiacea) under pack ice during the Antarctic
spring of 1985. J. Crustacean Biol. 7, 437±48.
O'Brien, D.P. (1988) Surface schooling behaviour of the coastal krill Nyctiphanes australis
(Crustacea: Euphausiacea) off Tasmania, Australia. Mar. Ecol. Prog. Ser. 42, 219±33.
O'Brien, D.P. (1989) Analysis of the internal arrangement of individuals within crustacean
aggregations (Euphausiacea: Mysidacea). J. Exp. Mar. Biol. Ecol. 128, 1±30.
Onsrud, M.S.R. & Kaartvedt, S. (1998) Diel vertical migration of the krill Meganyctiphanes
norvegica in relation to physical environment, food and predators. Mar. Ecol. Prog. Ser.
171, 209±219.
Pauly, T., Higginbottom, I., Nicol, S. & de la Mare, W.K. (1996) Results of a hydroacoustic
survey of Antarctic krill populations in CCAMLR Division 58.4.1 carried out in January to
April, 1996. CCAMLR, Hobart, WG-EMM-96/28.
Pitcher, T.J., Magurran, A.E. & Winfield, I.J. (1982) Fish in larger shoals find food faster.
Behav. Ecol. Sociobiol. 10, 149±51.
Ragulin, A.G. (1969) Underwater observations of krill. Trudy VNIRO 66, 231±34.
Ricketts, C., Watkins, J.L., Morris, D.J., Buchholz, F. & Priddle, J. (1992) An assessment of
the biological and acoustic characteristics of swarms of Antarctic krill. Deep-Sea Res. 39,
359±71.
Ringelberg, J. (1995) Changes in light-intensity and diel vertical migration: a comparison of
marine and freshwater environments. J. Mar. Biol. Ass. UK 75, 15±25.
Ritz, D.A. (1994) Social aggregation in pelagic invertebrates. Adv. mar. Biol. 30, 155±216.
102
Krill: Biology, Ecology and Fisheries
Roe, H.S.J. (1983) Vertical distributions of euphausiids and fish in relation to light intensity in
the Northeastern Atlantic. Mar. Biol. 77, 287±98.
Ross, R.M., Quetin, L.B. & Lascara, C.M. (1996) Distribution of Antarctic krill and dominant
zooplankton west of the Antarctic Peninsula. In Foundations for Ecological Research West
of the Antarctic Peninsula (R.M. Ross, L.B. Quetin & E.E. Hofmann, eds), Antarctic
Research Series, 70, pp. 199±217. American Geophysical Union, Washington.
Siegel, V. & Kalinowski, J. (1994) Krill demography and small-scale processes: a review. In
Southern Ocean ecology: the BIOMASS perspective (S.Z. El-Sayed, ed.), pp. 145±63.
Cambridge University Press, Cambridge.
Siegel, V. & Loeb, V. (1995) Recruitment of Antarctic krill Euphausia superba and possible
causes for its variability. Mar. Ecol. Prog. Ser. 123, 45±56.
Simard, Y. & Lavoie, D. (1999) The rich krill aggregation of the Saguenay±St. Lawrence
Marine Park: hydroacoustic and geostatistical biomass estimates, structure, variability, and
significance for whales. Can. J. Fish. Aquat. Sci. 56, 1182±97.
Simard, Y., Ladurantaye, R. de & Therriault, J.-C. (1986) Aggregation of euphausiids along a
coastal shelf in an upwelling environment. Mar. Ecol. Prog. Ser. 32, 203±215.
Strand, S.W. & Hamner, W.M. (1990) Schooling behavior of Antarctic krill (Euphausia
superba) in laboratory aquaria: reactions to chemical and visual stimuli. Mar. Biol. 106,
355±59.
Stretch, J.J., Hamner, P.P., Hamner, W.M., Michel, W.C., Cook, J. & Sullivan, C.W. (1988)
Foraging behavior of Antarctic krill Euphausia superba on sea ice microalgae. Mar. Ecol.
Prog. Ser. 44, 131±39.
Tarling, G.A., Buchholz, F. & Matthews, J.B.L. (1999) The effect of a lunar eclipse on the
vertical migration behaviour of Meganyctiphanes norvegica (Crustacea: Euphausiacea) in
the Ligurian Sea. J. Plankton Res. 21, 1475±88.
Tarling, G.A., Matthews, J.B.L., Saborowski, R. & Buchholz, F. (1998) Vertical migratory
behaviour of the euphausiid, Meganyctiphanes norvegica, and its dispersion in the Kattegat
Channel. Hydrobiologia 376, 331±41.
Trathan, P.N., Everson, I., Miller, D.G.M., Watkins, J.L. & Murphy, E.J. (1995) Krill biomass
in the Atlantic. Nature 373, 201±202.
Watkins, J.L. (1986) Variations in the size of Antarctic krill, Euphausia superba Dana, in
small swarms. Mar. Ecol. Prog. Ser. 31, 67±73.
Watkins, J.L. & Murray, A.W.A. (1998) Layers of Antarctic krill, Euphausia superba: are
they just long krill swarms? Mar. Biol. 131, 237±47.
Watkins, J.L., Buchholz, F., Priddle, J., Morris, D.J. & Ricketts, C. (1992) Variation in
reproductive status of Antarctic krill swarms; evidence for a size-related sorting mechanism? Mar. Ecol. Prog. Ser. 82, 163±74.
Watkins, J.L., Morris, D.J., Ricketts, C. & Priddle, J. (1986) Differences between swarms of
Antarctic krill and some implications for sampling krill populations. Mar. Biol. 93, 137±46.
Witek, Z., Kalinowski, J. & Grelowski, A. (1988) Formation of Antarctic krill concentrations
in relation to hydrodynamic processes and social behavior. In Antarctic Ocean and
Resources Variability (D. Sahrhage, ed.), pp. 237±44. Springer-Verlag, Berlin.
Chapter 5
Population Parameters
Volker Siegel and Stephen Nicol
5.1
Background on life history information
Introduction
Euphausiids, or krill, are important elements in many coastal and oceanic pelagic
ecosystems and are targeted by commercial fisheries in a number of areas (Nicol &
Endo, 1997). Krill populations fluctuate in response to environmental changes
which as a result of fishing activities may have severe ecosystem implications.
Understanding the population response to changes and managing the effect of
fisheries requires information on the key population parameters of the species
involved as well as their natural variability. Euphausiids are found in a range of
habitats from the tropics to the poles but are most important in the ecosystems of
polar and temperate waters which experience considerable seasonality (Mauchline
& Fisher, 1969; Mauchline, 1980). Large seasonal environmental changes make it
necessary for krill to develop strategies in growth, reproduction or overwintering to
survive poor conditions and maintain stock stability. Of particular importance in
attempting to clarify these strategies at both the individual and population level is
the knowledge of basic body components, e.g. length, total wet mass, dry mass, and
the formulation of length±weight relationships. This information may allow the
development of indices of `condition' and forms the backbone for production and
energy budget determinations (Falk-Petersen & Hopkins, 1981).
Information on how krill species have adjusted to long winters with low temperatures, little food or light is critical to an understanding not only of the species
survival but of the entire polar food webs. Other matters complicating the situation
are seasonal and longer-term environmental changes in the system. If long-term
climate variability is present, the question arises whether krill population dynamics
will show enduring effects and how serious these effects are on the stock structure
and stock abundance±biomass. Furthermore these changes in krill stocks are
obviously critical for dependent higher trophic levels in the system with krill as the
central component of the food web and at the basis of the trophic chain. These
interactions are discussed at length in Chapter 7.
Management of emerging and established krill fisheries also requires information
on a number of population parameters. The largest fishery, that for Antarctic krill,
103
104
Krill: Biology, Ecology and Fisheries
has the most sophisticated management regime and the generalised yield model
used to set precautionary limits on the fishery requires primary information on the
biomass, recruitment and growth rate of the krill population (Constable & de la
Mare, 1996). This model is applicable to other fisheries and there have been indications that this approach may be adopted by other krill fisheries (Head, 1997) and
this will require the collection of biological information on the species being harvested.
The present contribution summarises published, but widely scattered, basic
information on key population parameters of post-larval krill: age, growth, mortality and recruitment, particularly for those krill species which are of commercial
interest. The concentration is on those aspects of the life history which are critical to
understanding processes associated with fisheries management or ecological processes. Information on other species is included where it helps to illustrate a point of
interest. Much of the earlier material on these topics has been reviewed in two
volumes (Mauchline & Fisher, 1969; Mauchline, 1980) and no attempt has been
made to further summarise this literature.
Methodologies
Key population parameters of euphausiids have been investigated by studies at sea
and in the laboratory. Both of these approaches have advantages and disadvantages
and much of what we know of the better studied species of krill comes from a
combination of both types of study. Field-based methods have used net-collected
samples of krill, usually from surveys which have used an unbiased sampling
approach, to estimate the population parameters of krill based on morphometrics
and classification of sex and maturity stages. Most often, these have been infrequent
or annual surveys although there have been some studies which have used monthly
(or shorter) sampling intervals. Only a few studies have used very short sampling
intervals to obtain estimates of changes in population parameters over a few days.
There are a number of methodological problems with net-based surveys and these
are discussed by Watkins, Chapter 4. Nets are thought to be selective when dealing
with euphausiids and are subject to avoidance (Everson & Bone, 1986). Fine-mesh
nets may collect younger stages well but are thought to be evaded by the larger
animals. There are, however, no regularly used techniques which can provide
samples of krill for analysis or laboratory study other than nets. Samples from
predators have been used to provide time series through a season, but there are
large gaps in our knowledge as to what extent these samples are biased, e.g. by krill
availability in a localised area or selectivity by the predator and its potential
variability due to behavioural changes.
There are also uncertainties regarding the degree to which net surveys provide a
representative sample of the population. Because krill are pelagic, often schooling
organisms which can occur patchily in huge abundances, the ability of nets to
adequately obtain a representative sample of the population has been questioned
Population Parameters
105
(Miller & Hampton, 1989). High sampling density is the best guarantee that a
synoptic picture of a krill population can be obtained; however, ensuring that the
same population is sampled over time is more problematic. In some species of krill,
there are thought to be resident populations in fjords which can be repeatedly
sampled (Falk-Petersen & Hopkins, 1981). For many open ocean or shelf species,
however, there is no certainty that the same population is being sampled from one
survey to the next. Indeed, in the Antarctic, the prevailing theory is that Antarctic
krill are subject to flux, being moved quite rapidly from one area to another by
large-scale ocean currents (Murphy et al., 1998). To assess changes in Antarctic krill
populations, therefore, the assumption must be made that population processes
inferred from small-scale surveys are affected by processes on a very much larger
scale such that the sampling intervals are reflective of the residence time of krill in
the larger-scale area.
Because of their small size, pelagic habit and high abundance, and their frequent
moulting, krill are not amenable to mark-recapture studies so field-based experiments on the growth and maturity of individuals have not been attempted. Consequently, experimental investigations on krill have been restricted to laboratory
studies and such studies into population parameters of krill have been confined to a
small number of species. This is partly because of the difficulties that have been
encountered in maintaining euphausiids in the laboratory for significant periods of
time (Baker, 1963; Lasker & Theilacker, 1965; Komaki, 1966; Mackintosh, 1967;
Clarke, 1976; Ikeda et al., 1980; Ross, 1981; Miller et al., 1983). Only E. superba
(Ikeda, 1984; Ikeda, 1987a, b), and E. pacifica (Ross, 1982a; Ross, 1982b) have been
reared from eggs to adults. E. crystallorophias (Ikeda, 1986), Nematoscelis difficilis
(Gopalkrishnan, 1973) and Euphausia lucens (Pilar, 1984) have been reared through
a significant portion of their life history. The paucity of information on living
animals has meant that for most euphausiid species estimates of growth rates,
development rates and longevity are based on analysis of preserved specimens.
Laboratory studies have a number of inherent problems which encourage caution
when extrapolating their results to the wild population. Euphausiids are pelagic
organisms so their collection and confinement in experimental apparatus imposes
obvious stresses. Additionally, many species of krill, including all of those which
have been experimentally studied, are aggregating or schooling organisms and their
experimental manipulation in isolation may inflict unrealistic conditions on them.
Both field-based and laboratory-based studies have a number of caveats associated with them. A further problem that occurs when integrating the information
that arises from both types of study lies in assessing which type of information is
more useful in particular circumstances. These sorts of conflicts can be resolved in
future by the use of more realistic laboratory-based experiments, the use of fieldbased experiments and the directed study of certain areas where the evidence from
the two approaches are contradictory. An attempt has been made to highlight those
areas which are a priority for future studies and where aspects of krill biology can be
advanced by concerted study.
106
5.2
Krill: Biology, Ecology and Fisheries
Age determination and population age structure
Introduction
Like all crustaceans krill possess an exoskeleton. The process of growth requires the
shedding of this relatively inflexible structure so size increases are stepped rather
than linear. Euphausiids are unusual amongst crustaceans because they moult
regularly during their entire life even when they have reached maturity and have
ceased growing and even if they are deprived of food (Nicol & Stolp, 1991). This
regular periodic loss of their hard structures means that there are no known growth
rings or incremental changes equivalent to those found in vertebrate bones, scales or
otoliths. Consequently age determination of individuals or age structure of populations of all species of euphausiid has been somewhat fraught with difficulty.
Up until the late 1970s age determination of all species of euphausiids was
determined from length±frequency analyses. The experimental finding that krill
shrink, if starved (Ikeda & Dixon, 1982), raised suggestions that length-based ageing
methods were possibly subject to more error than had been previously thought. All
animals have the ability to shrink; however, Antarctic krill were shown to reduce
their size in sub-optimal food conditions by moulting to a smaller size and to lose
their sexual characteristics over time (Thomas & Ikeda, 1987). This finding introduced further uncertainties into the use of measures of length, combined with an
assessment of the sexual maturity, as definitive indicators of age. These results
precipitated a debate on ageing and mortality of krill that is yet to be resolved
(Nicol, 1990). Further results of experimental studies on Antarctic krill showed
evidence of shrinkage by populations of Antarctic krill in the laboratory (Nicol et
al., 1991) as well as by individuals. Groups of krill with polymodal size distributions
tend to contract to a unimodal size distribution at the lower end of the original size
range when kept in aquaria under low food conditions (Nicol et al., 1991; Sun, 1997).
This alteration of the shape of the size spectrum may come about through the differing food requirements of the smaller and larger animals ± the small krill receiving
a maintenance ration but the larger ones suffering from a sub-optimal diet and
hence shrinking. This differential effect is that most likely to cause difficulties for
length±frequency analyses if it occurs under natural conditions.
Although the issue of shrinkage has been discussed primarily with regard to
Antarctic krill the questions raised through these experimental studies are equally
applicable to other species. Shrinkage has been shown in the laboratory for five
species of euphausiids: E. pacifica (Lasker, 1966), E. superba (Ikeda & Dixon, 1982),
M. norvegica (Buchholz, 1985), Thysanoessa inermis (Dalpadado & Ikeda, 1989),
Nyctiphanes australis (Hosie & Ritz, 1989). There is experimental evidence that all
species of krill so far examined can shrink in the laboratory so the cautions
expressed over the use of length±frequency analysis of Antarctic krill are probably
applicable to other species.
Studies, using length±frequency analysis, have also revealed apparent reduction
Population Parameters
107
in mean or modal size which can be interpreted as winter shrinkage in particular size
classes or in whole populations (Hollingshead & Corey, 1974; Bamstedt, 1976;
Kulka & Corey, 1978; Falk-Petersen & Hopkins, 1981; Hopkins et al., 1984; FalkPetersen, 1985; Asthorsson, 1990; Dalpadado & Skjoldal, 1991; Bollens et al., 1992;
Dalpadado & Skjoldal, 1995; Dalpadado & Skjoldal, 1996). The observed decrease
in mean sizes of these populations has also been attributed to overwinter mortality
among the older (larger) animals or to immigration or emigration. However,
reduction in mean length during winter has not been reported to the extent that the
modal size structure of the length±frequency distribution was distorted sufficiently
to disrupt the modal progression or distribution mixture analysis for these species
(Falk-Petersen, 1985).
Ageing from length-frequency analysis
Despite reservations raised as a consequence of the results of experimental studies,
most analyses of krill populations continue to rely on length±frequency analysis to
provide measures of age and growth. Length±frequency analysis has two major
assumptions: that the same population is being measured at each sampling and that
the animals grow and mature in some predictable fashion, but both assumptions still
have to be verified for pelagic organisms like euphausiids.
Only in a few isolated cases has there been direct evidence that the same local
population of krill has been followed throughout a seasonal cycle. Several studies of
the seasonal population biology of temperate euphausiids have been conducted in
enclosed seas or fjords, in which the same population can be sampled repeatedly
(Hollingshead & Corey, 1974; Kulka & Corey, 1978; Hopkins et al., 1984; FalkPetersen, 1985; Asthorsson, 1990; Bollens et al., 1992).
In short-lived species the length±frequency distributions may ideally show a
unimodal size distribution, representing a spawning event and the growth of the
population over the season. Polymodal length±frequency distributions occur in
warm or temperate water species, when discrete spawning pulses during the course
of the year are separated by growth, e.g. E. pacifica off southern California (Brinton,
1976), Nyctiphanes couchii in the Bay of Biscay (Gros & Cochard, 1978), T. longicaudata (Lindley, 1978), N. australis off New Zealand (Jillett, 1971). These seasonal
spawning events have a cohort structure and can often be followed over the entire
life span of the species, which rarely extends over more than one year. In the case of
continuous spawning over the year no modes would be obvious from the size distribution and cohorts are no longer detectable. Examples for these non-structured
length±frequency distributions are tropical species such as Nyctiphanes simplex and
Euphausia eximia off the west coast of Baja California (Gomez, 1995).
Cold water species from Antarctic or subarctic regions also show polymodal
length±frequency distributions. Most of these species have a single relatively short
breeding season every year, e.g. E. superba (Marr, 1962; Mauchline & Fisher, 1969);
E. triacantha (Baker, 1959); M. norvegica (Einarsson, 1945). Each spawning event
108
Krill: Biology, Ecology and Fisheries
then represents a cohort or age group and for younger size classes leads to a clear
mode in their length±frequency distribution. Therefore, in such cases it can be
concluded that the polymodal patterns indicate a multi-year life cycle of the species.
If there are no size dependent net selection factors (methodological biases) or prey
size selection factors by predators (natural mortality influences) one can generally
assume that the length±frequency distribution of a cohort represents a normal
distribution.
The maximum lifespan of krill has been estimated by simple modal progression
analysis, which follows the sequential progression of the peaks in length±frequency
samples. This works for species with few cohorts and a relatively fast growth rate, for
example: T. inermis, T. raschii and E. pacifica (Einarsson, 1945; Bollens et al., 1992).
If there are too many cohorts produced in one year for a short-lived species, or if too
many cohorts of a long-living species survive for several years and the growth rate of
the older animals slows down substantially, then the modes become less and less
distinguishable, because of overlapping normal distributions.
Since size distribution mixtures of species with a multi-year life cycle consist of a
number of normal distributions, statistical methods have been developed to facilitate the separation of mixture components into normal distributions. Graphical
analysis on probability paper (Harding, 1949) gives the mean length of the size
groups (Matthews, 1973). For older age classes this method is highly subjective and
does not allow an estimation of the reliability of the analyses.
An interactive computer program called MIX reduced the subjectivity; the
results include estimates of the proportions of the single components, the mean
length, and the standard deviation of the mean as well as the standard errors of
all estimates (Macdonald & Pitcher, 1979). Furthermore, a goodness of fit
between the observed and the expected distribution mixture is given. A similar
approach was described by (Bhattacharya, 1967) and applied to T. inermis populations (Dalpadado & Skjoldal, 1995). Even if the maximum number of age
groups for long-lived species like E. superba cannot be fully resolved, due to
overlapping oldest size groups, these methods can at least give the minimum
number of age groups in the population.
For quantitative analyses of distribution mixtures (e.g. recruitment estimates)
length±frequency data may be inadequate, because, in euphausiids they are not
normally distributed and length±frequency data represent standardised, but pooled,
averaged data. This could strongly bias the quantitative analysis of the different age
classes. For this reason the statistical properties of the distribution of density values
have been analysed and the MLMIX method was established to create the length±
density distribution (de la Mare, 1994). He also modified the Macdonald and Pitcher
least-square approach of fitting the data to the maximum likelihood method. As a
result the quantitative proportions as well as the mean length of the distribution
mixture components can be calculated including a statistical test for the goodness of
fit to the data. This technique has yet to be applied to populations of other species of
euphausiids.
Population Parameters
109
Lifespan and size at age
Despite the issues raised above, methods using analyses of length±frequency data
have found wide application for the determination of age and growth in krill.
Statistical methods applied to distribution mixtures were first successfully used for
Antarctic krill by Aseev (1984) (graphical method on probability paper) and Siegel
(1986, 1987) applying the Macdonald and Pitcher approach. These authors found a
much longer lifespan for this species than the previously suggested 2-year life cycle
(Rudd, 1932; Marr, 1962; Mauchline & Fisher, 1969) (Table 5.1). Analyses from other
studies and areas in principal confirmed the results of a multi-year life cycle (Hosie et
al., 1988; Pakhomov 1995a; Wang et al., 1995). E. superba populations around the
continent show very similar results on length-at-age data with two deviations (Table
5.2). Firstly, krill at high latitudes like the inner Weddell Sea region seem to be
smaller than those from the more seasonal pack-ice zone. Secondly, in these high
latitude areas, the oldest krill are age 4+, whereas they reach at least age 5+ in the
Atlantic and in parts of the Indian Ocean sector (Siegel, 1987; Hosie et al., 1988). This
difference was first noted by Marr (1962) and juvenile Weddell Sea krill indeed differ
in mean length from more northerly waters. However, older age classes in the
Weddell Sea are very similar in size to Antarctic Peninsula and Indian Ocean krill.
Second, the largest age classes in the Indian Ocean sector are absent (Hosie et al.,
1988; Pakhomov 1995a), either indicating a statistical problem of fitting the distribution mixture to the observed data, or indicating a missing age class. This may be
because these very old age classes strongly overlap in size and occur only in low
abundance. However, the high variability in mean length of the 4+ group (Hosie et al.,
1988), and the very similar mean length values for all younger classes may support the
conclusion that, at least in some years, age class 5+ animals have been attributed
erroneously to 4+, even though older animals did exist in the population. The detailed
work of Lu & Wang (1996) on the separation of distribution mixtures clearly
demonstrates that a high increase in the significance level of the fit is possible when
assuming the existence of a 5+ age group instead of a maximum age of 4+. It is also
evident that the mean length-at-age for the younger age classes is very stable and the
results are very robust whether the procedure includes or excludes a 5+ age group. In
this case only the mean length of the 4+ group is obviously affected.
Whether there are age classes beyond 5+ for the Antarctic krill, those cannot be
resolved with the present methods because there is too much overlap in the distribution of size groups. However, the age group 5+ is already below 1% of the stock
size, so that even older age groups would not contribute substantially to the
population or strongly influence population dynamic parameters.
Other euphausiid species age structure
The sympatric species T. inermis and T. raschii have a wide boreal to low arctic
distribution, in the Atlantic from temperate waters off Nova Scotia and the North
inermis
inermis
inermis
inermis
inermis
inermis
inermis
inermis
inermis
inermis
inermis
inermis
inermis
inermis
inermis
inermis
Thysanoessa
Thysanoessa
Thysanoessa
Thysanoessa
Thysanoessa
Thysanoessa
Thysanoessa
Thysanoessa
Thysanoessa
Thysanoessa
Thysanoessa
Thysanoessa
Thysanoessa
Thysanoessa
Thysanoessa
Thysanoessa
Meganyctiphanes
norvegica
Meganyctiphanes
norvegica
Meganyctiphanes
norvegica
N North Sea, S Norway
boreal
subarctic
arctic
Scotland
S Barents Sea
N Norway
E Greenland
raschii
raschii
raschii
raschii
raschii
raschii
raschii
raschii
Thysanoessa
Thysanoessa
Thysanoessa
Thysanoessa
Thysanoessa
Thysanoessa
Thysanoessa
Thysanoessa
1+
2+
2+
Scotland
W Norway
1+
1+
1+
2+
2+
2+
2+
2+
2+
2+
2 yr 3 mo
3+
3+
3+
3+
2+
1+
2+
2+
2+
2+
2+
2 yr 3 mo
3+
Max
age
Mediterranean
N North Sea, S Norway
N North Sea
W Norway
boreal
subarctic
arctic
Scotland
S Barents Sea
N Norway
Canada, Nova Scotia
N Norway
Barents Sea
Greenland
E Greenland
N Pacific, NE Aleutians
N Pacific, SE Aleutians
Geogr. region
Species
2+
few 1 + F, few 1 + M
2+
1+
2+
few 1 + F, most 1 + M
2+
3+
3+?
2+
1+
2+
3+
3+?
2+
1+
1+
1+
1+
2+
no spawning
1+
2+
2+
1+
1+
1+
2+
1+
1+
2+
no spawning
Age at 1st
spawning
1+
1+
2+
2+
Age at maturity
Mauchline, 1960, Mauchline & Fischer,
1969
Sager & Sammler, 1984
Lindley, 1982
Lindley, 1980
Jones et al., 1967
JoÈrgensen & Matthews, 1975
Einarsson, 1945
Einarsson, 1945
Einarsson, 1945
Mauchline, 1966
Drobysheva, 1987
Berkes, 1976
Kulka & Corey, 1978
Falk-Petersen & Hopkins, 1981
Dalpadado & Skjoldal, 1991, 1995, 1996
Lindley, 1980
Wiborg, 1968
Nemoto, 1957
Nemoto, 1957
Lindley, 1980
Einarsson, 1945
Einarsson, 1945
Einarsson, 1945
Mauchline, 1966
Drobysheva, 1987
Falk-Petersen & Hopkins, 1981
Lindley, 1980
Reference
Table 5.1 Longevity and age at maturation for different euphausiid species in different geographical regions; F = females, M = males, R = recruits.
110
Krill: Biology, Ecology and Fisheries
pacifica
pacifica
pacifica
pacifica
pacifica
pacifica
pacifica
pacifica
pacifica
pacifica
pacifica
pacifica
superba
superba
superba
superba
superba
superba
superba
superba
superba
superba
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Meganyctiphanes
norvegica
Meganyctiphanes
norvegica
Meganyctiphanes
norvegica
Meganyctiphanes
norvegica
2+
2 yr 6 mo
W Norway
N Norway
Antarctic, S. Georgia
Antarctic Peninsula
Antarctic Peninsula
Antarctic, Indian Ocean
Antarctic, Indian Ocean
Antarctic, Indian Ocean
Antarctic, Indian Ocean
Antarctic, Indian Ocean
Antarctic, Indian Ocean
Antarctic, Scotia Sea
4+
5+
5+
5+
4+
4+
5+
5+
6+
6±7 yr
1 yr
1+
6±8 mo spring R
10±13 mo fall R
15 mo
19±22 mo
2+
2+
2+
2+
2 + F, 1 + M
21 mo
2+
Skagerrak
Oregon
Oregon, Washington
S California
S California
NE Japan
Canada BC
NW Pacific
NW Pacific, Kamchatka
NW Pacific
N Pacific, Aleutians
N Japan
SW Japan
2+
Skagerrak
0 = larv; 1 + = juv
3 + f, 3 + M
3+
1+
1+
3 + F, 3 + M
3+
1+
1+
1+
1+
2* F, 3 + M
4 mo
7 mo
4 mo
7 mo
2 + F, 3 + M
1 yr
no spawning
1+
1+
1 yr
1
1+
Ivanov, 1970
Siegel, 1986, 1987, Siegel & Loeb, 1994
Kawaguchi et al., 1997
Aseev, 1984
Hosie et al., 1988
Pakhomov, 1995a
Wang et al., 1995
Lu & Wang, 1996
Ettershank, 1983, 1985
Rosenberg et al., 1986
Smiles & Pearcy, 1971
Brodeur & Pearcy, 1992
Brinton, 1976
Brinton, 1976
Iguchi et al., 1993
Nicol & Endo, 1997
Brinton, 1976
Ponomareva, 1963
Nemoto, 1957
Iguchi & Ikeda, 1995
Endo, 1981, cf. Nicol & Endo, 1997
Iguchi et al., 1993
Falk-Petersen & Hopkins, 1981
Lindley, 1980
Boysen & Buchholz, 1984
Einarsson, 1945
Population Parameters
111
112
Krill: Biology, Ecology and Fisheries
Table 5.2 Mean length-at-age data (TL in mm) for different euphausiid species in different
regions.
Species
Geogr. region
Season
Thysanoessa raschii
Thysanoessa inermis
Thysanoessa inermis
Thysanoessa inermis
Thysanoessa inermis
Thysanoessa inermis
Thysanoessa inermis
Thysanoessa inermis
N Atlantic boreal
Barents Sea
Barents Sea
Barents Sea
Barents Sea
Barents Sea
Barents Sea
Barents Sea
spawning season
June
Oct
Jan
May
Oct
Jan
end of growth season
Meganyctiphanes
Meganyctiphanes
Meganyctiphanes
Meganyctiphanes
Meganyctiphanes
Meganyctiphanes
Skagerrak
Scotland
Norway
N Atlantic
N Atlantic
Kattegat
winter, range of size classes
winter, range of size classes
winter, range of size classes
winter (mean TL)
winter (means for diff. years)
winter (mean TL)
norvegica
norvegica
norvegica
norvegica
norvegica
norvegica
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
pacifica
pacifica
pacifica
pacifica
pacifica
pacifica
pacifica
pacifica
pacifica
pacifica
S California
S California
S California
S California
S California
Oregon
S Japan
NW Pacific
NW Pacific
NE Japan
Feb. cohort after 7 months
June cohort after 10 months
1st spawning
2nd spawning
3rd spawning
Sep. cohort after 13 months
0 group in winter, 1+ at 21 mo
spring
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
superba
superba
superba
superba
superba
superba
superba
superba
superba
superba
superba
Antarctic Peninsula
Antarctic Peninsula
Antarctic Peninsula
Antarctic Peninsula
Antarctic, Weddell Sea
Antarctic, Weddell Sea
Antarctic, Indian Ocean
Antarctic, Indian Ocean
Antarctic, Indian Ocean
Antarctic, Indian Ocean
Antarctic, Indian Ocean
Feb±March
March±April
May/June
October, spring
Jan±March, summer
Oct±Nov. winter/spring
Dec±March
Dec±Feb
Jan±Feb
March±April
0 group in winter, 1+ summer
Sea to almost high arctic waters off Greenland and the northern Barents Sea. In the
Pacific their distribution ranges from north of 408N in the western part and north of
508N in the eastern part extending into the Bering and Beaufort Seas. Due to their
wide latitudinal range these species show an interesting succession of population age
structures (Table 5.1).
At the limit of their southern distribution T. inermis and T. raschii attain a
maximum age of one year, dying after the first spawning season. In boreal waters
around Scotland, Iceland, western Norway and Nova Scotia maximum age of the
species increases to two years. Age-at-maturity and first spawning still occurs at the
Population Parameters
Age 0
12.1
11.7
9.5
9.9
10.5
13±25
24±26
14±27
18
20
19
19
11.5
16
20
10
10±12
3±8.5
8.0
6.7
12.7
1.0±2.0
1.2±7.3
1+
2+
13
13.4
19.8
20.3
13.1
12.8
12.6
13.9
21
20.8
22.5
24.5
22.1
17.1
16.6
18.5
27±35
36±37
29±34
25
20±23
31
42±48
43±47
32
31±36
36
22
20
13±14
17±18
16±18
29.0±31.8
30.0
28.7
30.5
20.8±25.2
26.4
29.6±32.2
19.0±30.8
22.3±28.4
25.7±30.6
28.4
3+
4+
5+
1991
1991
1991
1991
1991
1991
1995
Poulsen, 1926
Mauchline, 1960
Wiborg, 1971
Einarsson, 1945
Lindley, 1982
Boysen & Buchholz, 1984
Brinton, 1976
Brinton, 1976
Brinton, 1976
Brinton, 1976
Brinton, 1976
Smiles & Pearcy, 1971
Iguchi et al., 1993
Ponomareva, 1963
Nemoto, 1957
Taki & Ogishima, 1997
19
22
38.0±41.3
39.4
36.0
40.0
39.2
39.3
37.4±38.3
34.2±40.9
39.0±41.8
32.8±40.7
41.8
Reference
Einarsson, 1945
Dalpadado & Skjoldal,
Dalpadado & Skjoldal,
Dalpadado & Skjoldal,
Dalpadado & Skjoldal,
Dalpadado & Skjoldal,
Dalpadado & Skjoldal,
Dalpadado & Skjoldal,
24.2
22.1
21.5
22.7
113
45.5±47.6
44.5
43.6
46.0
44.5
44.8
43.4±44.8
41.2±48.8
44.2±46.4
40.3±48.3
46.4
51.0±52.3
50.0
50.3
55.5±57.1
54.2
50.1
47.8±50.7
51.0±54.9
49.2±51.5
49.2±55.0
50.3
52.4±55.4
54.8
54.8
Siegel, 1986
Siegel, 1987
Siegel, 1987
Siegel, 1987
Siegel, 1987
Siegel, 1987
Aseev, 1984
Hosie et al., 1988
Wang et al., 1995
Pakhomov, 1995a
Lu & Wang, 1996
age of 1+ and part of the population survives for another year and spawns for a
second time. Off northern Norway these species further demonstrate their
developmental plasticity. Maximum age is still 2+, and most males become mature
as 1-year-olds; however, only very few females spawn at the age of 1+ and most
spawn once at the age of 2+ (Falk-Petersen & Hopkins, 1981). This trend continues
in the Barents Sea and for the Pacific in the southeastern Aleutian region, where
individual krill reach a maximum age of 3+ (Nemoto, 1957; Lindley, 1980). Here
they mature and spawn at 2+, however, the number of second time spawners at the
age of 3+ must be very low, because this oldest age class is rarely reported (see Table
114
Krill: Biology, Ecology and Fisheries
5.1.). Greenland and the northeastern Aleutians are probably the most environmentally extreme locations, because both species require three years to reach
maturity and only spawn for a single season. Further north in high arctic waters they
still mature, but spawning does not occur and the populations must be replaced by
advection from the south.
Well-documented mean length-at-age data for T. inermis and T. raschii are only
available from Barents Sea studies (Table 5.2). Although Einarsson (Einarsson,
1945) did not identify the 3+ age group, his mean length-at-age data for Thysanoessa
krill in the North Atlantic are similar to those from the Barents Sea (Dalpadado &
Skjoldal 1991; Dalpadado & Skjoldal 1995). The maximum age of these species
changes with latitude, but not the length-at-age. In spite of this there are records
indicating that Thysanoessa grow to a larger maximum size close to their northern
distribution limits (Einarsson, 1945).
M. norvegica has a wide distribution range in the Atlantic and its southern range
extends into the Mediterranean Sea, where it has a maximum life expectancy of one
year and a single spawning season during its life cycle. Off western Norway M.
norvegica lives to the age of 2+ and, as in the Skagerrak and off Scotland, spawns for
the first time at 1+ and performs a possible repeat spawning as a 2-year-old. Unlike
the Thysanoessa species, the maximum age of M. norvegica does not increase off
northern Norway, and in the southern Barents Sea, it is still documented as 2 years
and 6 months, when it does not appear to spawn there (Falk-Petersen & Hopkins,
1981; Drobysheva, 1987). Although the mean length-at-age for most areas seems to
be consistent, older age classes seem to be greater in mean length in Scottish and
Norwegian waters (Table 5.2). Mean length-at-age data from the North Atlantic for
different years show only minor inter-annual differences (Lindley, 1982).
E. pacifica also shows large variation in maximum age. It experiences the shortest
life span of 6 to 8 months along the coast of California which is at its start of the
southern limit of its distribution. After 4 months the spring recruits reach maturity
and spawn for the first time with the potential of repeat spawning every 2 months. In
the same area autumn cohorts survive to the age of 10 to 13 months. However,
maximum size of the cohorts is very similar (Brinton, 1976). In Japanese waters
maximum age increases to 15 to 21 months (Iguchi et al., 1993). Off the coast of
Oregon and Washington longevity is similar to that in Japanese waters and mean
length-at-age for 0+ and 1+ krill is comparable on both sides of the Pacific (Smiles &
Pearcy, 1971).
At higher latitudes (Aleutians, Kamchatka) E. pacifica matures and spawns for
the first time at 1+ and part of the population survives the following winter and
spawns again at 2+ . Although maximum size of the species seems to be similar for
most of the distribution range, the mean length-at-age in high latitude regions is
obviously smaller. The mean length of 1+ krill off Oregon and Japan is 20 to 22 mm,
whereas in northern regions it is 13 to 18 mm and reaches 19 to 22 mm mean length
at 2+.
In principal, maximum age and age at sexual maturity increase with latitude. For
Population Parameters
115
E. pacifica and Thysanoessa spp, however, an intermediate step can be observed in
mid latitudes. The maximum age of Thysanoessa does not change, but most males
become mature at 1+, whereas only very few females spawn at the age of 1+ (see
above). For E. pacifica both sexes reach maturity and spawn as 1+, but only females
survive for another season and spawn a second time as 2+ (Endo, 1981; Nicol &
Endo, 1997) indicating a flexible response to latitudinal changes. Further, E.
superba shows a marked difference in sexual maturation; in this case females mature
and spawn first at the age of 2+, while male maturation is retarded by one year to
age class 3+ (Siegel & Loeb, 1994).
Other techniques for estimating krill age and longevity
Age pigments
The prospect of Antarctic krill shrinking during winter prompted investigations into
the use of fluorescent age pigment accumulation as an indicator of age (Ettershank,
1983). Fluorescent age pigments, including lipofuscin, are substances which accumulate with time in the cells of all organisms so far examined (Hammer & Braum,
1988). They are thought to be the products of cellular lipid peroxidation processes;
accumulating at a rate dependent on the metabolic rate of the animals. They are
fluorescent, emitting light at 420±470 nm when excited by wavelengths of 340±
370 nm. Although initial studies using fluorescent age-pigments as an indicator of age
held promise (Ettershank, 1983), subsequent investigations demonstrated
methodological problems which raised doubts about the reliability of the technique
used (Nicol, 1987; Mullin & Brooks, 1988). Studies examining fluorescence of soluble
age-pigment in other organisms have yielded equivocal results (Hirche & Anger,
1987; Hammer & Braum, 1988; Hill & Radkte, 1988; Vernet et al., 1988) but the utility
of this technique for ageing E. superba is yet to be proven. Laboratory studies with E.
superba indicated that some soluble fluorescent age-pigments accumulated with age
in a population that shrank significantly in length over the period of a year in captivity
(Nicol et al., 1991). Studies with captive populations of other species of crustaceans
have shown significant accumulation of age-pigments when examined using image
analysis and fluorescence microscopy (Sheehy et al., 1994; Wahle et al., 1996) and this
technique might prove more reliable than the solvent extraction method if applied to
euphausiids. It seems unlikely that the rather time-consuming methodologies for
quantifying levels of fluorescent age-pigments will supplant the more convenient and
rapid length-frequency analysis. More likely, they may become a useful way of
ground truthing the results of more conventional techniques.
Multiple morphometrics
Studies on E. superba and T. macrura have attempted to use multiple morphometrics to obtain more information on krill population age structure than can be
116
Krill: Biology, Ecology and Fisheries
obtained by conventional length±frequency analysis (Farber-Lorda, 1990; FarberLorda, 1991; Farber-Lorda, 1994). Although this approach yielded some promising
preliminary results in separating out groups of adults of both sexes, it has not been
adopted in other studies.
Eyeballs
A novel approach to ageing in E. superba has involved utilising some measure of the
size of the eyeball relative to the body size (Sun et al., 1995; Sun & Wang, 1995a, b,
1996; Sun, 1997). The premise of this approach is that the compound eye is a
complex structure and would be conserved if the animal is shrinking. Consequently,
the eye of an adult krill which had shrunk would be relatively larger when compared
to the eye of a juvenile of the same size which had been growing. The initial results
of this approach showed that laboratory-shrunk Antarctic krill had relatively larger
eyes (from crystalline cone counts) than might be predicted from their overall body
length. Further, the number of crystalline cones in the eyes of krill obtained from a
sample caught in spring (the period following the greatest supposed food scarcity
and hence the greatest likelihood of shrinkage) was higher than might be predicted
for their size and resembled those of shrunken laboratory krill (Sun et al., 1995).
Subsequently, eyeball diameter has been shown to be a good substitute for crystalline cone counts and eyeball diameter appears to remain constant when krill
shrink in the laboratory (Sun & Wang, 1996; Sun, 1997). Although this technique
has shown promise there are no recent reports of attempts to apply it to field
samples from other areas. One important practical outcome of this method is that it
appears possible to use the crystalline cone number or eyeball diameter to distinguish older krill from new recruits. This is sufficient information to estimate the key
parameters of recruitment and mortality (de la Mare 1994a, b).
Maintenance
The potential age of Antarctic krill has been demonstrated by laboratory studies
which have indicated that they can live in captivity for over 6 years giving a total
longevity of at least 7±8 years (Ikeda, 1987). There have been no reported studies of
maintaining other species of krill for extended periods and for most species, the
estimated life-span is derived from size frequency analysis.
Growth
Growth rates
Krill growth rates have been estimated in a number of ways. The most frequently
used method is the analysis of sequential length±frequency plots from samples
obtained throughout a seasonal cycle. Experimental studies have provided useful
Population Parameters
117
comparative measures of laboratory growth rates for larvae, juveniles and adults.
Finally, a combination of field and laboratory measurements ± the instantaneous
growth rate (IGR) methodology (Nicol et al., 1993; Quetin et al., 1994) has produced
some recent results for Antarctic krill and this technique may be applicable to other
species. Daily growth rates are often difficult to compare not only because of different methods of measurement, but also because growth is a function of size, age,
maturation, feeding condition and environmental condition. Furthermore, growth
in many species has a strong seasonal component. Most laboratory experiments
have utilised constant environmental conditions which may lead to unrealistically
high or low growth rates.
Field estimates of summer growth rates in E. superba populations have been
calculated from time series of length±frequency data, (Miller & Hampton, 1989).
Only two studies have been reported which track changes in the mean length of krill
populations over short periods of time. Kanda et al. (1982) observed a change in
mean size of a population of E. superba from 41.13 to 41.98 mm in 12 days corresponding to a daily change of 0.063 mm. Over a 30-day moult cycle this would
represent a size change at moult of ~ 4.6% which is consistent with laboratory
measurements. Clarke & Morris (1983) measured the modal length change of a
population of E. superba sampled over a 6-day period and observed a 2 mm increase
corresponding to a 0.33 mm day±1 increase. The krill in their study were ~ 30 mm
long so for a 30-day moult cycle the percentage increase at moult would be 33%,
considerably higher than other estimates. Other studies on Antarctic E. superba
have used data obtained from much more widely spaced sampling intervals which
may introduce a degree of error since it is not known whether the same population is
being sampled over time. McClatchie (1988) predicted growth rates of between
0.041 to 0.125 mm day±1 for adult and juvenile E. superba in Admiralty Bay during
summer (~ 4.1±12.5% increase in length per moult, again assuming a 30-day moult
cycle) which are similar figures to those derived from the IGR measurements
(Quetin & Ross, 1991). Growth rates of E. superba have been calculated from
modal peaks in the length frequencies from the South West Indian Ocean sector
(Farber-Lorda, 1994). The estimated growth rate for each of the identified year
classes was: 1+: 0.12±0.133, 2+: 0.077±0.08, 3+: 0.048±0.057, 4+: 0.03±0.034, 5+:
0.019±0.022 ± in mm day±1 ± equivalent to up to 18% per moult.
The growth rate of other species of krill has been estimated mostly from length±
frequency analyses. The data for M. norvegica indicate that growth is similar in the
Skagerrak and Kattegat and in North Atlantic waters (Poulsen, 1926; Einarsson,
1945; Lindley, 1982; Boysen & Buchholz, 1984). Although growth is only part of the
energy budget, daily growth rates in g C day±1 showed that most of the carbon goes
into body growth, the carbon net growth efficiency is highest in the E. pacifica
calyptopis stage (60 to 74%), but decreases to 10% in the adults. About 3 to 4% goes
into moults and 10 to 18% into reproduction during the adult stage, which is on
average more than used for growth at this age (Ross 1982a, b). This may explain how
T. inermis and T. raschii can reach a larger maximum size but at the same time
118
Krill: Biology, Ecology and Fisheries
postpone maturation and spawning by one or even two years in subarctic regions
(Einarsson, 1945; Lindley, 1980; Dalpadado & Skjoldal, 1991, 1995). Even for these
species the high arctic waters are so unfavourable that they no longer reproduce,
leading to the conclusion that there are no true arctic euphausiid species (Einarsson,
1945; Dalpadado & Skjoldal, 1996; Weigmann-Haass, 1997).
Field-derived estimates of growth rates of E. superba indicate that daily growth
rates decrease with age. Larvae during the first year of life grow at 0.13 to
0.16 mm d±1, juveniles of age 1+ at 0.12 to 0.148 mm d±1, while older animals grow
only at 0.025 to 0.07 mm d±1 (see Table 5.3). Growth parameter estimates for krill
sampled during the `Discovery' investigations were determined by Rosenberg et al.
(1986) using the von-Bertalanffy growth function:
Lt ˆ L1 1 ÿ eÿK…tÿt0 †
Where Lt is the length at age t, L1 is the asymptotic length, K is the coefficient of
catabolism and t0 is the theoretical origin of the equation.
The k values for the yearly as well as the monthly data were highly consistent. However, the estimated yearly k values of the juvenile krill were approximately 25% of those for the high growth season indicating that maximum
growth of E. superba (up to 0.179 mm day±1) occurs over little more than three
months (Rosenberg et al., 1986). This conclusion is similar to that derived from
analyses of growth data using a seasonal VBGF which concluded that krill
growth in the Antarctic Peninsula area is probably confined to the period from
late October to the end of January/early February (Siegel, 1982). These data
fit the theoretical growth curves established by Mauchline (1980), if the annual
growth period is limited to 90 days.
In the Bering Sea growth rate for postlarval T. raschii and T. inermis is about 2%
dry body weight per day, but is higher in warmer years, often > 5% (Table 5.3) when
they grow larger (Smith, 1991). In the North Sea T. inermis grew at a rate of 1.5%
dry body weight per day during March to June (Lindley & Williams, 1980). Growth
rates were highest (almost four times as high as the average) early in the season
during and shortly after the spring bloom. In northern Norway in spring and summer
T. inermis and T. raschii grew at an average rate of 1.2% and 1.7% dry body weight
per day, respectively, (Hopkins et al., 1984). Lipid synthesis in T. inermis is about
0.2% body weight per day, which obviously is a high proportion of the total daily
growth rate (Smith, 1991).
E. pacifica is short-lived in Californian waters, about 8 months for winter and
early spring cohorts, but growth can exceed 3 to 3.5 mm/month. Late spring and
summer cohorts have a life expectancy of 12 months and growth rates are similar to
the winter cohorts except for the slowing down in growth to 1.5 mm/month from
October at the start of autumn (Brinton, 1976). Longer living animals (15 to 21
months) in Japanese waters show a similar rapid growth during spring, March to
May, (Nishikawa et al., 1995) but zero growth from November to March (Taki &
Iceland
Scotland
Spain
Scotland
N Iceland/Greenland
S Iceland
N Norway
N Norway
N Norway
N Norway
North Sea, mean
North Sea, before bloom
North Sea, during bloom
North Sea, mean
Antarctic Peninsula
Washington, 8±128C
Washington, 8±128C
Washington, 8±128C
Washington, 8±128C
Antarctic Peninsula
Meganyctiphanes norvegica
Meganyctiphanes norvegica
Meganyctiphanes norvegica
Thysanoessa raschii
Thysanoessa inermis
Thysanoessa inermis
Thysanoessa raschii, 1st + 2nd winter
Thysanoessa raschii, summer
Thysanoessa inermis, 1st + 2nd winter
Thysanoessa inermis, summer
Thysanoessa inermis, adult
Thysanoessa inermis, adult
Thysanoessa inermis, adult
Thysanoessa inermis, larvae
Euphausia superba, larvae
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
pacifica, calyptopis
pacifica, furcilia
pacifica, juvenile
pacifica, adult
superba, C1 to F5 larvae
Geogr. region
Species
0.29±1.43
1.37±5.41
4.08±11.95
7.79±33.83
2.93±13.61
1.5
8.3
0.11
70.09
0.040
0.093
0.126
0.238
0.378
0.383
0 group
0.230
0.36
Growth rate in ug C/day
Growth rate in ug DW/day
107.7
44.1
152.5
70.03
1.02
Growth rate in % WW/day
70.21
1.53
0.68
0.127
0.110
0.184
0.31
0.476
Growth rate in mg WW/day
1+ group
older groups
0.132
0.101
0.155
0.274
0.427
0.383
life-time
Continued
Ross, 1982
Ross, 1982
Ross, 1982
Ross, 1982
Huntley & Brinton, 1991
Lindley & Williams, 1980
Lindley & Williams, 1980
Lindley & Williams, 1980
Lindley & Williams, 1980
Huntley & Brinton, 1991
Hopkins et al., 1984
Hopkins et al., 1984
Hopkins et al., 1984
Hopkins et al., 1984
Mauchline & Fischer, 1969
Mauchline & Fischer, 1969
Mauchline & Fischer, 1969
Mauchline & Fischer, 1969
Mauchline & Fischer, 1969
Mauchline & Fischer, 1969
References
Table 5.3 Growth rates of euphausiid species; IGR = instantaneous growth rates IGR* are measured in mm (or % of TL) per intermoult period
Bering Sea
Bering Sea
N Norway
N Norway
N Norway
N Norway
Oregon
S California
S California
S California
S California
Washington
Canada BC
Canada BC
Washington
Washington
Thysanoessa inermis, subadult F
Thysanoessa inermis, adult M
Thysanoessa raschii, 1st + 2nd winter
Thysanoessa raschii, summer
Thysanoessa inermis, 1st + 2nd winter
Thysanoessa inermis, summer
pacifica
pacifica, February cohort
pacifica, July cohort
pacifica, September cohort
pacifica, December cohort
pacifica
pacifica, spring
pacifica
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia pacifica, < 10 mm, summer
Euphausia pacifica, > 10 mm, summer
Sea
Sea
Sea
Sea
Sea
Sea
Sea
Sea
Bering
Bering
Bering
Bering
Bering
Bering
Bering
Bering
Thysanoessa
Thysanoessa
Thysanoessa
Thysanoessa
Thysanoessa
Thysanoessa
Thysanoessa
Thysanoessa
raschii, Nauplius 1
raschii, Calyptopis 1
raschii, subadult M
raschii, subadult F
raschii, adult M
raschii, adult F
raschii, calyptopis
raschii, subadult
Geogr. region
Species
Table 5.3 Continued
0.05±0.08
0.08±0.09
0.053±0.096
0.085
0.073
0.062
0.090
0.095
0.05±0.08
0.094
70.03
70.02
6.5±13.5
9.0±11.5
1.1±4.7
1.3±9.4
1.8±4.0
0.6±2.4
14.0
0 group
0.73
0.98
Growth rate in mm TL/day
70.05
1.19
70.29
1.74
1.0±6.0
1.3
3.0
Growth rate in % DW/day
Growth rate in mg WW/day
1+ group
older groups
0.038
0.065
life-time
Smiles & Pearcy, 1971
Brinton, 1976
Brinton, 1976
Brinton, 1976
Brinton, 1976
Bollens et al., 1992
Fulton & LeBrasseur, 1984
Heath, 1977 cf. Bollens
et al., 1992
Cooney, 1971 cf. Bollens
et al., 1992
Cooney, 1971 cf. Bollens
et al., 1992
Hopkins et al., 1984
Hopkins et al., 1984
Hopkins et al., 1984
Hopkins et al., 1984
Smith, 1991
Smith, 1991
Smith, 1991
Smith, 1991
Smith, 1991
Smith, 1991
Smith, 1991
Smith, 1991
Vidal & Smith, 1986
Vidal & Smith, 1986
References
N Japan
N Japan
N Japan
S Japan
Antarctic, experimental
Antarctic Peninsula, IGR*
Antarctic Peninsula, IGR*
Antarctic Peninsula, IGR*
Antarctic, Indian Oc.
IGR*
Antarctic, theoretical
Antarctic, theoretical
Antarctic Peninsula
Antarctic Peninsula
Antarctic Peninsula
Antarctic, Atlantic
Antarctic, Atlantic
Antarctic Peninsula
Antarctic, Indian Ocean
Antarctic, Indian Ocean
Antarctic, Atlantic
Antarctic, Indian Ocean
Antarctic, Indian Ocean
pacifica, larvae, summer
pacifica, winter
pacifica, summer
pacifica, larvae, summer
superba
superba, 22±44 mm
superba
superba, juvenile
superba, larvae, summer
superba, larvae, winter
superba, 32 mm
superba, 40 mm
superba, 40 mm
superba, larvae, winter
superba, juveniles/adults
superba
superba
superba
superba, 24±36 mm
superba
superba, larvae, winter
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
respe
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
season
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
superba, growth season
superba, mean
superba, summer
superba, winter
superba, larvae, summer
superba, larvae, winter
superba, max. growth
Canada BC
Euphausia pacifica, winter
experimental
experimental
experimental
experimental
experimental
experimental
Canada BC
Euphausia pacifica, summer
Antarctic,
Antarctic,
Antarctic,
Antarctic,
Antarctic,
Antarctic,
Canada BC
Euphausia pacifica
0.047
0.163
0.070
0.130
0.017
0.06±0.11
0.02
0.100
0.133
0.000
0.001±0.004
0.066
0.095
0.120±0.133
0.133
0.109
0.120
0.148
2.4 to 9.5%
0.047
0.056
0.047
0.075
0.039±0.078
0.070
0.058
0.019±0.080
0.025
0.010±0.048
0.3/7.3%
0.083±0.156
0.056
70.04 to 70.812
0.070
0.033
0.130
0.032
0.024±0.044
0.0345
Siegel, 1986
Kanda et al., 1982
Pakhomov, 1995a
Clarke & Morris, 1983
Hosie et al., 1988
Hosie & Stolp, 1989
Mauchline, 1980
Mauchline, 1980
McClatchie, 1988
McClatchie, 1988
Huntley & Brinton, 1991
Daly, 1990
Rosenberg et al., 1986
Murano et al., 1979
Ikeda et al., 1985
Poleck & Denys, 1982
Ikeda & Thomas, 1987a, b
Ikeda, 1984
Elias, 1990, cf. Quetin
et al., 1994
Buchholz et al., 1989
Quetin & Ross, 1991
Quetin & Ross, 1991
Ross & Quetin, 1991
Nicol et al., 1993
Hulsizer, 1971 cf. Bollens
et al., 1992
Hulsizer, 1971 cf. Bollens
et al., 1992
Hulsizer, 1971 cf. Bollens
et al., 1992
Nishikawa et al., 1995
Taki & Ogishima, 1997
Taki & Ogishima, 1997
Iguchi et al., 1993
122
Krill: Biology, Ecology and Fisheries
Ogishima, 1997). In the higher latitudes of the NW Pacific E. pacifica shows a slower
growth rate and longer life cycle compared to lower latitudes (California and
Oregon) (Nemoto, 1957; Ponomareva, 1963). Growth rates are almost twice as high
in the south and the number of generations per year also decreases from California
to the north. In Japanese waters E. pacifica reaches the same maximum size (Iguchi
et al., 1993). Rapid growth of E. pacifica off Oregon and California may be related to
the high productivity of the upwelling events and the lack of large seasonal temperature fluctuations (Smiles & Pearcy, 1971). Under experimental conditions
increasing temperatures did indeed shorten intermoult periods in E. pacifica, but
growth increments did not change (Iguchi & Ikeda, 1995). Optimum temperature
for growth in this species is thought to be about 10±138C.
Growth in euphausiids probably is not temperature dependent, but related to
food availability (Clarke & Peck, 1991). Daily growth rates can be quite high at the
height of the production season which indicates that a substantial (but not necessarily perfect) degree of temperature compensation in growth rate occurs. If growth
in polar species is affected by seasonal resource limitation rather than temperature,
some predictions can be made:
(1)
(2)
Growth will be seasonal in herbivores and less so in carnivores under the same
environmental conditions.
Maximum growth rates will be comparable between polar species and similar
species from warmer waters (Clarke & Peck, 1991).
Good examples for the first prediction are the euphausiids E. superba, T. inermis
and T. raschii which show high growth rates modulated by a strong seasonal pattern
(e.g. Rosenberg et al., 1986; Siegel, 1986, 1987; Dalpadado & Skjoldal, 1996; FalkPetersen et al., 1981; Falk-Petersen & Hopkins, 1981). The species M. norvegica and
Euphausia triacantha are known to feed throughout the winter as omnivores or even
pure carnivores and although growth slows down at the end of the winter these
species grow continuously in length with a much less pronounced seasonal growth
pattern (Boysen & Buchholz, 1984, Falk-Petersen, 1985; Baker, 1959; Siegel, 1987).
Interestingly, the data for M. norvegica also indicate that growth is similar for different geographical regions like the Skagerrak and Kattegat and North Atlantic
(Poulsen, 1926; Einarsson, 1945; Lindley, 1982; Boysen & Buchholz, 1984).
Similar maximum growth rates have been reported from a range of species from
polar and more temperate waters. E. superba daily growth rates estimated from field
samples during the summer growth period (Table 5.3), can be as high as those
theoretically predicted by Mauchline (1980) for an annual growth period of 90 days.
From these results it is apparent that the growth potential of the Antarctic krill E.
superba is comparable to other species like Thysanoessa spp or M. norvegica in
more temperate waters. It is not the potential growth rate in E. superba that is
slower during the summer growth period, but the shorter annual growth period in
polar latitudes that is responsible for lesser annual increase in size. Those results
Population Parameters
123
support the view that krill growth performance is more or less independent of
temperature within the range to which the species is adapted, but it is the food
resource supply that drives the growth performance. Similarly, the rapid growth of
T. inermis shows that the species must have a highly efficient biosynthesis despite
the low summer-water temperature. Fatty acid biosynthesis from cold-water Norway was triple that of specimens from temperate Scotland. From this Falk-Petersen
(1985) drew the conclusion that the constraints leading to an overall reduced annual
growth in high latitudes are not physico-chemical, but ecological.
Seasonality and winter stagnation
Parameter C of the seasonal von-Bertalanffy growth function (seasonal VBGF in
the form of Lt = L? (1 7 e± [k(t ± to) + CU/(2m) sin 2m (t ± tS)]) with ts as the starting point of
the sinus oscillation) describes the amplitude of growth oscillation which normally
ranges between 0 and 1; low values express a low seasonality in the growth amplitude, values near 1 indicate a stagnation of growth in winter, larger values mean a
high amplitude and possible degrowth in winter. Seasonal VBGF curves in some
cases show that growth of T. inermis ceases during winter (Dalpadado & Skjoldal,
1991, 1996) or between March and October for E. superba (Siegel, 1986) (Table 5.4).
The only seasonal VBGF curve for M. norvegica shows a C value that suggests a lack
of any seasonal stagnation in growth (Sager & Sammler, 1984).
M. norvegica is known to feed throughout the winter as a carnivore and although
growth slows down a little at the end of the winter, growth in length is continuous
(Boysen & Buchholz, 1984). Wet weight of individuals was relatively stable, while
dry weight decreased at the end of winter (Falk-Petersen, 1985). In the Mediterranean, there is a rapid increase in body size during the first year followed by a
tenfold reduction in growth rate in the second (Labat & Cuzin-Roudy, 1996). This
pattern was attributed to a concentration on somatic growth during the first year
and gonadal production during the second. The sizes of the two-year classes in the
Mediterranean are somewhat smaller than those found in the North Atlantic and
the North Sea despite the growth rates being similar. The patterns of growth for this
species are likely to vary considerably in this species because of its wide geographic
range extending over regimes with different productivity.
Laboratory estimates
The growth rates of E. superba derived from laboratory studies have been summarised by Buchholz (1991) and a range of positive growth increments up to 21%
per moult have been reported. Most of these laboratory growth studies have relied
on measurements of length taken from consecutively moulted exoskeletons which
gives a coarse-scale picture of growth over a whole moult cycle. One laboratory
study has used a version of the `instantaneous growth rate technique', measuring the
length of the krill and their newly produced exuviae (Buchholz et al., 1989) and an
30.19
60.0
63.8±64.2
61.0
61.3±63.3
56.6
61.3±65.1
British Columbia
Antarctic, Scotia Sea
Antarctic Peninsula
Antarctic Peninsula
Antarctic Peninsula
Antarctic Weddell Sea
Antarctic, N Indian
Ocean
Antarctic, S Indian
Ocean
59.1±60.1
40.5±41.2
W Norway
Euphausia superba
45.0
30.3
20.11
Barents Sea
Barents Sea
Barents Sea
Thysanoessa inermis
Thysanoessa inermis
Thysanoessa
longicaudata
Meganyctiphanes
norvegica
Euphausia pacifica
Euphausia superba
Euphausia superba
Euphausia superba
Euphausia superba
Euphausia superba
Euphausia superba
L inf
Geogr region
Species
0.466±0.467
0.0085
0.43±0.47
0.384±0.401
0.4728
0.354±0.478
0.532
0.417±0.440
0.751±0.776
0.16
0.53
0.56
70.027
0.88
0.20
1.09
70.002/70.044
70.004/70.106
0.141
0.06±0.155
0.072
70.044/70.101
70.049/70.057
70.41
70.29
70.27
0.959
0.47±0.51
2.80
71.55
1.47
VBGF parameters t = 1 year, 1* = per day
k
ts
to
C
sVBGF
t*
sVBGF
sVBGF
sVBGF
sVBGF
Pakhomov, 1995a
recalc. from Sager &
Sammler, 1984
Tanasichuk, 1998
Rosenberg et al., 1986
Siegel, 1986
Siegel, 1986
Siegel, 1987
Siegel, 1987
Pakhomov, 1995a
Dalpadado & Skjoldal, 1996
Dalpadado & Skjoldal, 1991
Dalpadado & Skjoldal, 1996
References
Table 5.4 Parameters of von Bertalanffy growth function. L inf = asymptotic length, k = growth constant, ts = starting point of oscillation, to = origin of growth curve, C
= amplitude of growth oscillation, sVBGF = seasonalised version
Population Parameters
125
average increase in length of 3.8% per moult was observed under conditions of high
food availability.
Laboratory growth rates of E. pacifica have been reported for individuals from
populations in widely separated parts of its range. Iguchi & Ikeda (1995) report
growth rates of 1.1±2.0% per moult with moulting rates of between 2 and 25 days.
Long-term food deprivation results in a relatively constant shrinkage rate of 0.033
mm per day (nearly 73.3% per moult cycle) in juvenile E. superba (Ikeda et al.,
1985). This figure is similar to the maximum rate of shrinkage demonstrated by
Antarctic krill in IGR experiments that have been allowed to proceed beyond five
days (Nicol et al., 1993). Thus under experimental conditions krill can apparently
achieve their maximal shrinkage rate within a matter of days of food limitation and,
in the laboratory, krill have survived periods of starvation-induced shrinkage for 211
days (Ikeda & Dixon, 1982).
Instantaneous growth rate (IGR)
A methodology for the measurement of IGR was outlined by Quetin & Ross (1991)
and their results demonstrated both shrinkage and growth in freshly caught krill.
This technique was further modified by Nicol et al. (1993) to examine the effects of
capture on the measurements of growth rate and to increase the sample size (Nicol
et al., 1999). This technique is beginning to provide information on the growth rate
of a wide range of sizes of krill which can be related to environmental conditions and
to the observed changes in size of the whole population (Ross et al., 2000).
The instantaneous growth rate technique relies on collecting live krill in good
condition. On board, the krill are kept in individual containers and are checked
regularly for moults during the five days following capture. During this period the
growth rate of krill appears to be stable and is thought to reflect the natural growth
rate; after five days the growth rate starts to decline and is thought to reflect the
effect of confinement in the laboratory (Nicol et al., 1993). The growth rate is
estimated from the difference in length of the uropods on the moults and the whole
post-moult krill. Since Antarctic krill moult every 20±30 days at 08C (Buchholz,
1991) only 1 in 20 to 1 in 30 krill will moult on any given day of the experiment. This
methodology therefore requires large numbers of animals of which only a small
fraction will be used for actual measurements of initial growth rates. The key to
obtaining greater numbers of growth rate measurements is to increase the number
of krill that can be maintained individually at one time (Nicol et al., 2000). Using
small (250 ml), perforated jars to hold individual krill in a large (1000 l) flowthrough tank permits a sample size of over 1000 krill per experiment, a much larger
sample size than has been the case in earlier studies.
The IGR technique has shown positive growth of adult E. superba during the late
summer and autumn with increases in body length at moulting of between 1.75 and
4.4% (Quetin & Ross, 1991) and, in winter, shrinkage of between 70.16 and
72.03% of body length per moult. The growth rates of Antarctic krill in the Indian
126
Krill: Biology, Ecology and Fisheries
Ocean sector of the Southern Ocean were measured in four summers (Nicol et al.,
1993) using the IGR technique. Mean growth rates ranged from 0.35 to 7.34% per
moult in adults and 2.42 to 9.05% in juveniles. The mean growth rates of adult and
juvenile krill differed between areas and between the different years of the investigation. When food was restricted under experimental conditions, individual krill
began to shrink immediately and mean population growth rates decreased gradually, becoming negative after as little as seven days. Populations of krill which
exhibited higher initial growth rates began to shrink later than those which had
initially been growing more slowly. IGR measurements on Antarctic krill in the
southeastern Indian Ocean sector using the modified technique with large sample
sizes (up to 1000 animals moulting per experiment) provided results which again
indicated that juvenile krill exhibit the fastest growth rates (a mean of 6% per
moult) (Nicol et al., 2000). Males were growing slightly slower than juveniles (mean
4.3%), early stage females were growing at a higher rate (mean 2.4%) whereas
gravid and spent females were growing very slowly (mean 0.9%). All stages were
growing at a rate that was similar to the rate predicted from demographic
measurements of the mean change in size of each of the maturity and sex groups
over the sampling period.
The IGR technique appears to offer a method which provides data on growth
rates that can be compared to field-derived methods using length frequency analysis
from the same populations. This will allow some of the conflicting results on krill
growth that have emerged from field and laboratory studies to be examined.
Although the technique has been developed for Antarctic krill it should be
applicable to other species which are robust enough to withstand capture and
experimental confinement. For smaller temperate or tropical species, the more
rapid moulting rates may actually enhance the sensitivity of this technique by
providing far greater sample sizes.
Shrinkage
The intense seasonality of Antarctic regions has led to investigations into how
Antarctic krill populations survive the winter. Due to the low food availability
and the low lipid content in E. superba Ikeda & Dixon (1982) assumed that feeding ceases during winter and their experimental results indicated that if this happens they will shrink. A reduction in metabolic rate could also be the mechanism
that permits krill to overwinter without feeding (Quetin & Ross, 1991). Body
shrinkage would then account for only 4% of the energy required. Both scenarios
are based on the assumption that krill do not feed during the winter but this may
not always be the case. Feeding on ice-algae has now been reported several times
(Daly & Macaulay, 1988; Marschall, 1988; Stretch et al., 1988; Lancraft et al.,
1991) and krill may also feed on zooplankton or protozoa (Siegel, 1988; Lancraft
et al., 1991; Hopkins & Torres, 1989; Siegel, 1989; Lancraft et al., 1991). Winter
feeding by E. superba is also indicated by observed C:N ratios and ammonium
Population Parameters
127
excretion rates (Huntley et al., 1994). Shrinkage, starvation and even reduced
metabolism of Antarctic krill may be unusual phenomena in winter under natural
conditions.
Winter is the period when food is most likely to be limiting and hence this is the
period when overall shrinkage of krill populations would be likely to occur. Daly
(1990) reported significant growth of larval E. superba during winter in the marginal
ice zone but there are few unequivocal data sets which demonstrate either growth or
shrinkage of adult or larval Antarctic krill over winter. Krill captured under sea-ice
over winter showed a decrease in mean size and a restriction in the size range of krill
present (Kawaguchi et al., 1986). They also found little evidence of feeding in the
water column, because there was only low chlorophyll a concentrations throughout
the winter and the oxygen consumption figures corresponded to those of starving,
shrinking krill (Ikeda & Dixon, 1982). E. superba in Admiralty Bay (Antarctic
Peninsula region) may show winter shrinkage (Ettershank, 1983) but this change
may have been a result of either water movements into and out of the bay bringing
krill of different origin and life history or of differential size-related mortality
(McClatchie et al., 1991).
Reduction in the mean size of E. superba populations at the end of the winter has
not been observed in the few studies that have been sampled in this time period, nor
has there been a reported reduced range of sizes in each adult age-group. If
shrinkage does occur over winter and it occurs in a way that preserves the order of
the lengths between age groups across the population, then the method of decomposition of length±frequency distributions into separate components of length for
each age class remains valid (de la Mare, 1994a, b).
Using the IGR technique individual Antarctic krill have been observed shrinking
during winter (Quetin & Ross, 1991) although it is not certain whether shrinkage is a
widespread or common enough phenomenon to significantly affect the mean size of
the population. The major question is how frequently (if ever) krill encounter food
shortage sufficient to engender shrinkage and do they do so for sufficient periods to
affect the size structure of the population. Only in situ studies in winter and spring
will be able to answer these questions.
A reduction in mean length of T. inermis, mainly of age class 2+ during winter, has
been reported (Dalpadado & Skjoldal, 1996) but the mean size of this age group
never became as small as the mean summer length of age group 1+ (which also
shrank in winter), so a critical overlap in the length distribution between these two
cohorts never occurred and the age±size groups remained identifiable.
Euphausiids are obviously capable of shrinkage but when and where this occurs is
a matter for further research. How such shrinkage affects the analysis of population
dynamics of krill depends on whether the shrinkage occurs in a systematic fashion,
preserving the year classes or whether it tends to merge year classes. In analyses of
euphausiid populations it should be recognised that they are capable of shrinking
and that this should be taken into account as a potential source of uncertainty in lifehistory models.
128
5.4
Krill: Biology, Ecology and Fisheries
Length±weight relationships
A number of studies have reported relationships between the various body-length
measurements and weight measures for euphausiids (Table 5.5). Measurements of
length±weight relationships are affected by preservation; M. norvegica lost about
15% of the fresh weight when stored in formalin or even 21% when preserved in
alcohol (Kulka & Corey, 1982). These losses appear to be species-specific. When T.
inermis was transferred from formalin into alcohol, specimens lost another 30% of
their wet weight; more than one would expect from figures obtained for M. norvegica. It is clear that practical problems also exist in the accuracy of weight
measurements of fresh specimens onboard vessels.
Length±weight relationships in male and female M. norvegica in the Kattegat show
seasonal differences (Boysen & Buchholz, 1984). Although males are heaviest in
spring and autumn, females are heavier than males in spring, but after spawning in
late summer (April to September) are lighter. However, they attain the same weight
as males in autumn. In western Norwegian fjords monthly differences in length±
weight relationships of M. norvegica are observed, but differences between males and
females are not apparent, although males are, on average, heavier in late summer
(Bamstedt, 1976). M. norvegica appears to be heavier in summer and autumn than in
winter and spring. Significant seasonal differences in length±weight relationships
were also reported from E. pacifica in Japanese waters (Iguchi et al., 1993).
In populations in Northern Norway T. inermis shows monthly differences in
length±weight relationships and length±lipid relationships and no differences in
length±protein, while T. raschii shows significant differences in all three regressions
(Falk-Petersen, 1985). The decrease in dry weight is first seen in December and lasts
to March and it indicates that a negative energy balance lasts from mid-winter until
spring. Interestingly, M. norvegica showed significant differences in monthly length
to wet-weight relationships; however, absolute values for dry weight and protein
contents did not significantly change with season. The differences in the various
relationships of these three species may reflect different overwintering strategies in
the subarctic environment. T. inermis builds up lipid reserves as an energy resource
for the winter, T. raschii partly utilises lipids stores but additionally feeds on detritus
whereas M. norvegica is carnivorous (Falk-Petersen, 1985).
Length±weight relationships for the Antarctic krill E. superba. have been summarised previously (Morris et al., 1988; Siegel, 1992b). We have condensed some of
this large amount of information into a few length±weight relationships to facilitate
their practical use (Table 5.5). The first general observation on these relationships is
that length±weight relationships do not differ between males and females during
spring (November), when animals are in the pre-spawning stage (IIIA), but still are
of low weight. Generally in December there is a significant increase in weight with a
simultaneous progression of krill maturity to gravid stages; males and females still
do not show significant differences in length±weight relationships. After spawning
(January to March) the weight of males and females drops immediately with
females losing significantly more weight than males. From this low level both sexes
Geogr. region
N Norway
Canada, Nova Scotia
Canada, Nova Scotia
Canada, Nova Scotia
Canada, Nova Scotia
Kattegat, February
Kattegat, February
Kattegat, May
Kattegat, May
Kattegat, August
Kattegat, August
Kattegat, November
Kattegat, November
Canada, St. Lawrence
Canada, St. Lawrence
Canada, St. Lawrence
N Norway, May
N Norway, May
N Norway, August
N Norway, August
N Norway, November
N Norway, November
N Norway, February
N Norway, February
N Norway, May
N Norway, May
N Norway, August
N Norway, August
Species
Thysanoessa inermis
Meganyctiphanes norvegica
Thysanoessa inermis
Meganyctiphanes norvegica
Thysanoessa inermis
Maganyctiphanes norvegica
Meganyctiphanes norvegica
Meganyctiphanes norvegica
Meganyctiphanes norvegica
Meganyctiphanes norvegica
Meganyctiphanes norvegica
Meganyctiphanes norvegica
Meganyctiphanes norvegica
Meganyctiphanes norvegica
Thysanoessa raschii
Thysanoessa inermis
Thysanoessa inermis
Thysanoessa inermis
Thysanoessa inermis
Thysanoessa inermis
Thysanoessa inermis
Thysanoessa inermis
Thysanoessa inermis
Thysanoessa inermis
Thysanoessa raschii
Thysanoessa raschii
Thysanoessa raschii
Thysanoessa raschii
0.00144
0.00036
0.00023
0.00019
0.00145
0.00219
0.00014
0.00009
0.00029
0.000717
0.000712
a
1.202
0.134
0.682
0.084
0.746
0.094
0.657
0.065
0.521
0.039
0.634
0.142
72.580
70.136
70.342
71.005
71.083
3.330
2.730
2.882
2.717
2.884
3.029
3.397
3.533
3.629
3.028
2.848
3.729
3.876
3.500
3.170
3.380
2.203
2.338
2.543
2.885
2.532
3.007
2.607
3.130
2.667
3.235
2.654
2.666
WW
WW
WW
DW
DW
DW
DW
DW
DW
DW
DW
DW
DW
DW
DW
DW
WW
DW
WW
DW
WW
DW
WW
DW
WW
DW
WW
DW
TL
CL
CL
CL
CL
TL male
TL female
TL male
TL female
TL male
TL female
TL male
TL female
TL
TL
TL
CL
CL
CL
CL
CL
CL
CL
CL
CL
CL
CL
CL
Length/Weight relationship parameters
log a
b
WW/DW
Length
Continued
Dalpadado & Sjkoldal, 1991
recalc. from Kulka & Corey, 1982
recalc. from Kulka & Corey, 1982
recalc. from Kulka & Corey, 1982
recalc. from Kulka & Corey, 1982
Boysen & Buchholz, 1984
Boysen & Buchholz, 1984
Boysen & Buchholz, 1984
Boysen & Buchholz, 1984
Boysen & Buchholz, 1984
Boysen & Buchholz, 1984
Boysen & Buchholz, 1984
Boysen & Buchholz, 1984
Sameoto, 1976
Sameoto, 1976
Sameoto, 1976
Falk-Petersen, 1985
Falk-Petersen, 1985
Falk-Petersen, 1985
Falk-Petersen, 1985
Falk-Petersen, 1985
Falk-Petersen, 1985
Falk-Petersen, 1985
Falk-Petersen, 1985
Falk-Petersen, 1985
Falk-Petersen, 1985
Falk-Petersen, 1985
Falk-Petersen, 1985
References
Table 5.5 Length±weight relationships for different euphausiids, weight in mg and length in mm; a = regression constant, b = regression coefficient; WW = wet weight,
DW = dry weight, TL = total length, CL = carapace length, M = male, F = female
Geogr. region
N Norway, November
N Norway, November
N Norway, February
N Norway, February
N Norway, May
N Norway, July
N Norway, November
N Norway, February
N Norway, all year
W Norway, February
W Norway, May
W Norway, August
W Norway, November
N Atlantic
N Atlantic
Japan, summer
Japan, winter
Japan
N Atlantic
Nova Scotia
W Norway, November
W Norway, December
W Norway, June
Bering Sea, March
Bering Sea, April
Bering Sea, May
Antarctic Peninsula, winter
Antarctic Peninsula, January
Ant. Pen. Nov. Pre-spawning
Ant. Pen. Nov. pre-spawning
Species
Thysanoessa raschii
Thysanoessa raschii
Thysanoessa raschii
Thysanoessa raschii
Meganyctiphanes norvegica
Meganyctiphanes norvegica
Meganyctiphanes norvegica
Meganyctiphanes norvegica
Meganyctiphanes norvegica
Meganyctiphanes norvegica
Meganyctiphanes norvegica
Meganyctiphanes norvegica
Meganyctiphanes norvegica
Thysanoessa inermis
Thysanoessa raschii
Euphausia pacifica
Euphausia pacifica
Euphausia pacifica
Thysanoessa inermis
Meganyctiphanes norvegica
Meganyctiphanes norvegica
Meganyctiphanes norvegica
Meganyctiphanes norvegica
Thysanoessa raschii
Thysanoessa raschii
Thysanoessa raschii
Euphausia superba
Euphausia superba
Euphausia superba, M
Euphausia superba, F
Table 5.5 Continued
0.00076
0.00105
0.00099
0.097
0.092
0.361
0.074
0.0013
0.0013
a
73.540
70.99
70.85
70.42
71.192
71.210
70.914
73.250
72.909
71.170
72.580
72.270
0.409
0.056
0.449
0.054
0.151
0.698
0.744
0.779
0.039
3.500
2.820
2.910
2.350
3.056
3.257
2.874
3.270
2.866
3.071
2.965
2.917
3.265
2.858
3.208
3.310
2.714
2.711
2.693
3.028
3.004
2.954
2.480
3.218
2.916
2.916
3.310
3.020
3.156
2.770
DW
DW
DW
DW
DW
DW
DW
DW
DW
DW
DW
WW
DW
WW
DW
WW
WW
WW
WW
DW
DW
DW
DW
DW
DW
DW
WW
WW
DW
WW
TL
TL
TL
TL
CL
CL
CL
TL
TL
TL
TL
CL
CL
CL
CL
CL
CL
CL
CL
CL
CL
CL
CL
CL
TL
TL
TL
TL
TL
CL
Length/Weight relationship parameters
log a
b
WW/DW
Length
Falk-Petersen, 1985
Falk-Petersen, 1985
Falk-Petersen, 1985
Falk-Petersen, 1985
Falk-Petersen, 1985
Falk-Petersen, 1985
Falk-Petersen, 1985
Falk-Petersen, 1985
Falk-Petersen, 1985
Bamstedt, 1976
Bamstedt, 1976
Bamstedt, 1976
Bamstedt, 1976
Lindley, 1978
Lindley, 1978
Iguchi et al., 1993
Iguchi et al., 1993
Iguchi & Ikeda, 1995
Matthews & Hestad, 1977
(cf. Mauchline, 1980)
Sameoto, 1976b
Barattelid & Matthews, 1978
Barattelid & Matthews, 1978
Barattelid & Matthews, 1978
Vidal & Smith, 1986
Vidal & Smith, 1986
Vidal & Smith, 1986
Huntley et al., 1994
Huntley et al., 1994
Siegel, 1992
Siegel, 1992
References
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
superba, M
superba, F
superba, M
superba, F
superba, M
superba, F
superba, M
superba, F
superba, M
superba, F
superba, M
superba, F
superba, M
superba, F
superba, M
superba, F
superba, M
superba, F
superba, M
superba, F
Ant. Pen., gravid
Ant. Pen., gravid
Ant. Pen., February, spent
Ant. Pen., February, spent
Ant. Pen., adult
Ant. Pen., gravid
Ant. Pen.
Ant. Pen., non-gravid
Ant. Pen.
Ant. Pen., gravid
Ant. Pen., adult
Ant. Pen., non-gravid
Ant. Pen. Nov. Pre-spawning
Ant. Pen. Nov. Pre-spawning
Ant. Pen., gravid
Ant. Pen., gravid
Ant. Pen., February, spent
Ant. Pen., February, spent
Ant. Pen., winter, resting stage
Ant. Pen., winter, resting stage
0.00019
0.00025
0.00009
0.00031
0.00401
0.00199
0.00238
0.00139
0.00613
0.00975
0.01729
0.01088
0.00315
0.0043
0.00083
0.00115
0.00111
0.00211
0.00328
0.00441
3.435
3.357
3.694
3.306
2.790
3.043
2.927
3.073
3.077
2.981
2.815
2.908
3.207
3.102
3.561
3.457
3.507
3.302
3.176
3.084
DW
DW
DW
DW
DW
DW
DW
DW
WW
WW
WW
WW
WW
WW
WW
WW
WW
WW
WW
WW
TL
TL
TL
TL
TL
TL
TL
TL
TL
TL
TL
TL
TL
TL
TL
TL
TL
TL
TL
TL
Siegel, 1986
Siegel, 1986
Siegel, 1986
Siegel, 1986
Morris et al.,
Morris et al.,
Morris et al.,
Morris et al.,
Morris et al.,
Morris et al.,
Morris et al.,
Morris et al.,
Siegel, 1989
Siegel, 1989
Siegel, 1986
Siegel, 1986
Siegel, 1986
Siegel, 1986
Siegel, 1989
Siegel, 1989
1988
1988
1988
1988
1988
1988
1988
1988
132
Krill: Biology, Ecology and Fisheries
further steadily decline in weight to a minimal and similar winter level (Siegel,
1989). In summary, this indicates that differences in male and female length±weight
relationships are observed during a very brief period immediately after spawning
(with males heavier than females) and a significant difference between the spawning
season (high) and the rest of the year (low).
Morris et al. (1988) detailed variability of length±weight relationships in E.
superba and they found that the moult cycle has little effect upon length±weight
relationships. In their opinion a simple division of krill into males and females
without considering maturity stages does not improve the precision of the length±
weight relationships. However, separating first gravid females from males or nongravid females or secondly adult males from other males would improve the precision of weight predictions.
Weight at length has been examined as a `condition index' for E. superba in
the South-west Indian Ocean (Farber-Lorda, 1994) using a complex allometric
morphometric relationship. He found no significant differences as a function of
sex or developmental stage in either total length and weight or carapace length.
The changes in the relationship between the carapace length and dry weight have
also been used as a `condition factor' to examine overwintering in E. superba
(Kawaguchi et al., 1986). They showed systematic decreases in both carapace
length and dry weight over winter and an increase in both of these measurements
in spring. A more simple approach has indicated regional differences in the
weight of juvenile Antarctic krill of the same length in the South-east Indian
Ocean (Nicol et al., 2000) but whether this technique is generally applicable as a
condition index is uncertain.
Based on the above observations, it can be recommended that for greatest utility:
(1)
(2)
(3)
(4)
5.5
During the summer spawning period separate length±weight relationships
should be applied for gravid females, non-gravid females, adult male and other
male krill.
During the post-spawning season separate length±weight relationships should
be applied for the same categories as under (1) and additionally for spent
females.
During winter and spring general length±weight relationships can be used
which should be on a monthly basis in the pre-spawning period, although
statistically the differences are not significantly different before the occurrence
of gravid stages.
Where data are available only on sex and size of the animals, during and
shortly after the spawning period, separate length±weight relationships should
be applied for male and female krill. A greater improvement in precision is
gained by considering maturity stages separately.
Natural mortality
There are very few quantitative estimates of natural mortality in the published
Population Parameters
133
literature on euphausiid species (Table 5.6). For non-Antarctic species changes in
abundance of single modes over a period of time are presented as a reduction in
relative frequencies. Mortality of E. pacifica larvae is high in Californian waters.
Brinton (1976) estimates that only 16% of the larvae survive per month. After the
short larval phase the mortality rate is lower, about 67% of the juveniles survive to
adults per month, but mortality increases again after spawning and adult survival is
only about 60% per month for the adults until the next spawning event.
Mortality rates have been reported for E. pacifica (Jarre-Teichmann, 1996) with
M = 3.0 y-1 off California, M = 8.7 y-1 for Oregon and 0.6 to 1.9 y-1 for British
Columbia. These differences are difficult to explain, although the maximum life
span is different in these areas. As expected the lowest mortality rate is seen for the
animals living for 19±22 months off British Columbia. However, stocks off Oregon
and California reach the same maximum age of about one year, but their life cycle is
different. One cohort is reproducing about three times during its life off California,
but seems to reproduce only once off Oregon (Table 5.6), for which a much lower
mortality rate would be expected. One possible explanation could be that the
mortality rate estimate from Oregon was derived from a period of strong El NinÄo,
which may have increased the mortality of the species beyond normal levels.
Mortality of Thysanoessa species in the Barents Sea is lowest in winter, when the
feeding pressure of the main predator stock (cod and capelin) are minimal. Age
group 0+ has the highest summer growth rate, but also the highest mortality rate
(98%). Age group 1+ shows a mortality rate of approximately 50%, which still
leaves enough spawners (2+) in the next season to sustain the population. After
spawning mortality is thought to increase rapidly so that it was suspected that hardly
any 2+ animals survive (Drobysheva, 1987). However, other studies have found a
mortality rate of 55% y-1 for 2+ aged animals of T. inermis (Lindley, 1980), and the
presence of the 3+ age group has been reported, although in very low numbers in the
Barents Sea (Dalpadado & Skjoldal, 1991; Dalpadado & Skjoldal, 1995; Dalpadado
& Skjoldal, 1996). In Scottish waters T. raschii shows a slightly higher mortality rate
(75%) for the 1-year-olds than in the Barents Sea (Mauchline & Fisher, 1969).
Estimates of mortality rates for E. superba have been carried out, using different
methods. The results vary between 0.38 and 5.5 for the postlarval population (Table
5.6). Some of these seriously underestimated krill longevity. Since maximum age is a
function of natural mortality, this underestimation of longevity led to the highest
values. The Beverton and Holt (1959) approach takes into account the size composition of the catches under the assumption that the population is fully represented
in the samples (Pakhomov, 1995b) but this is not always the case. If age classes are
under-represented by recruitment failure or by net selectivity for larger animals, or
if sampling is not covering the geographical range of all size groups (Siegel, 1988),
then mortality rates can be seriously biased.
Mortality is inversely related to the longevity of the individuals of the population
and hence with the growth constant parameter `k' of the VBGF. High growth rates
are usually associated with high mortality rates, and low growth rates with low
mortality rates, so that the ratio, M/k, is approximately constant for a given species.
Antarctic, Scotia Sea
Antarctic Peninsula
Antarctic,
Antarctic,
Antarctic,
Antarctic,
Antarctic,
Antarctic,
Antarctic,
Antarctic,
Euphausia superba
Euphausia superba
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Euphausia
Ocean
Ocean
Ocean
Ocean
Ocean
Ocean
Ocean
Ocean
Antarctic, Scotia Sea
Euphausia superba
Indian
Indian
Indian
Indian
Indian
Indian
Indian
Indian
California
California
California
California
Oregon
Brit. Columbia
Barents Sea
Barents Sea
N Atlantic
NW Atlantic
Scotland
Scotland
Scotland
Antarctic, Indian Ocean
Antarctic, Scotia Sea
Antarctic, Scotia Sea
Antarctic Peninsula
Antarctic Peninsula
Antarctic Peninsula
Antarctic Peninsula
Antarctic, Scotia Sea
Antarctic, Scotia Sea
Euphausia pacifica
Euphausia pacifica
Euphausia pacifica
Euphausia pacifica
Euphausia pacifica
Euphausia pacifica
Thysanoessa spp
Thysanoessa spp
Thysanoessa inermis
Thysanoessa raschii
Thysanoessa raschii
Meganyctiphanes norvegica
Meganyctiphanes norvegica
Euphausia superba
Euphausia superba
Euphausia superba
Euphausia superba
Euphausia superba
Euphausia superba
Euphausia superba
Euphausia superba
Euphausia superba
superba
superba
superba
superba
superba
superba
superba
superba
Geogr. region
Species
postlarvae
postlarvae
postlarvae
year 1
year 2
year 3
year 4
year 5
postlarvae
postlarvae
postlarvae
< 2 year olds
2 and 3 year olds
postlarvae
postlarvae
postlarvae
postlarvae
postlarvae
postlarvae
larvae 0+
juvenile, 0+
adult, 0+
lifetime
lifetime
lifetime
0+
1+
1+
1+
1+
1+
2+
Age group or
develop. stage
0.52±0.57
0.72±0.87
0.76±2.92
1.09
0.65
0.57
0.77
1.54
0.38±1.22
0.75±1.13
0.45±0.65
1.83
0.40
0.51
3.00
8.70
0.60±1.90
3.90
0.69
0.80
0.29±1.2
1.39
0.69
2.30±3.00
5.50
2.31
0.51
0.78±1.17
0.88±0.96
0.66±0.92
0.94±0.99
0.81±1.35
0.5
Mortality M
y
y
y
y
y
y
y
y
y
y
y
mo
mo
mo
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
Per
y or mo
Pauly, VBGF: M; max
age: M
Richter & Efanov
Alverson & Carney
Beverton & Holt
Zikov & Slepokurov
Zikov & Slepokurov
Zikov & Slepokurov
Zikov & Slepokurov
Zikov & Slepokurov
Edmondson's
size group progression
size group progression
linearised catch curve
linearised catch curve
Alagajara, max age: M
Pauly, VBGF: M
VBGF K: M
1-cumulative length
frequency
length-depend.
predation curve
Method
Pakhomov, 1995a
Pakhomov, 1995a
Pakhomov, 1995a
Pakhomov, 1995a
Pakhomov, 1995a
Pakhomov, 1995a
Pakhomov, 1995a
Pakhomov, 1995a
Maklygin, 1992 in
CCAMLR Rep
Basson, 1994
Basson & Beddington, 1989
calc. from Brinton, 1976
calc. from Brinton, 1976
calc. from Brinton, 1976
cf. Jarre-Teichmann, 1996
cf. Jarre-Teichmann, 1996
cf. Jarre-Teichmann, 1996
calc. from Drobysheva, 1987
calc. from Drobysheva, 1987
Lindley, 1980
Lindley, 1980
Mauchline & Fisher, 1969
Mauchline & Fisher, 1969
Mauchline & Fisher, 1969
Kawakami & Doi, 1979
Brinton & Townsend, 1984
Brinton & Townsend, 1984
Siegel, 1986
Siegel, 1992a
Siegel, 1992b
Siegel, 1986
Priddle et al., 1988
Basson & Beddington, 1989
References
Table 5.6 Natural mortality rates (M) of euphausiids. As indicated under References in some cases M had to be calculated from data mentioned in these papers; y =
year, mo = month.
Population Parameters
135
For a large range of fish species, the ratio of M/k lies between 1.5 and 2.5 . Preliminary estimates for Euphausiids range from M/k = 1.5 to 2.5 based on the
regression equation of Pauly (1980) and between 0.83 and 1.67 for E. superba
obtained from the maximum age calculations (Basson, 1994). The range of M values
for E. superba from both approaches is 0.38 to 1.22.
Reviews of the methodology used in earlier studies concluded that the realistic
values of natural mortality for the postlarval life cycle may range between M = 0.66
and 1.35 (Siegel, 1992a; Pakhomov, 1995b). It is also likely that M varies between
geographical regions and interannually, and it is known to vary during the lifetime
of a species as already indicated above for T. inermis. Possible differences in mortality can be attributed to differing predator demands, as well as short-term variability or long-term trends of environmental parameters. There may also be
differences in natural mortality of different developmental stages or age groups.
Larval mortality, however, is not of direct concern for management; the basic
requirement is the estimate of M for those age classes typically affected by the
fishery.
Using the Zikov & Slepokurov (1982) approach Pakhomov (1995b) estimated
that the natural mortality rate for E. superba during the first year is relatively high
(M = 1.11 to 1.12), it decreases thereafter during the maturation period (M = 0.52 to
0.65) and increases slowly for older age groups, reaching a maximum during the last
year of life (M = 1.29 to 2.41). These results strongly confirm the dependence of
natural mortality on age. Similar studies in different regions and different years are
certainly needed, to evaluate possibilities of interannual or geographical variation in
mortality rates of certain age groups and consequently year classes, leading finally
also to variability in krill recruitment processes.
5.6
Recruitment variation
Recruitment of euphausiids has only recently been estimated following the introduction of a technique for E. superba (de la Mare, 1994a,b). For some other species
there are only descriptive comments on the success of single or few cohorts.
Recruitment of E. pacifica in Californian waters occurs after about 30 days, when
the larvae enter the juvenile phase. At least four generations are generally reported
each year. Due to the short life-span and the few cohorts, the maximum stock size of
this species is reached immediately after successful recruitment of a single cohort. In
general there is no spawning stock-recruitment relationship. In most years highest
recruitment is observed for the spring and summer cohorts, while autumn and
winter cohorts show low recruitment in these waters (Brinton, 1976). On the other
hand the July to December recruits live longer (10±13 months) than the shorterliving February to June recruits (6±8 months).
While up to four or more E. pacifica cohorts recruit off California every year, there
is only one large pulse of larvae recruiting in spring off the coast of Washington and a
distinct but less abundant cohort in late summer (Bollens et al., 1992). The reduced
136
Krill: Biology, Ecology and Fisheries
level of inferred spawning in summer is discussed by these authors as a result of the
much reduced phytoplankton level at this time of the year, although E. pacifica is
known to show omnivory. There is apparently no recruitment during winter as is the
case off California. Low temperatures cannot be responsible, because reproduction
of this species occurs in the northwestern Pacific at even lower temperatures.
Another possibility could be that survival of overwintering adult E. pacifica is low due
to predation in spring and summer and only few 1+ adults remain to spawn in late
summer. At the same time only few 0-aged animals are able to mature fast enough to
spawn during their first year of life (Ross et al., 1982).
In protected fjords of British Columbia recruitment occurs after around 47 days,
which is later than off California (Tanasichuk, 1998). In contrast to Washington
waters there is evidence for 4 to 6 cohorts recruiting every year with generally two
less successful spawning events in winter with no changes in spawning patterns and
frequencies between pre-post-ENSO and ENSO years. Maximum E. pacifica
abundance in El NinÄo years, however, was related to exceptional recruitment success (Tanasichuk, 1998). Interestingly this author also reports an inverse stockrecruitment relationship and suggests that high parental stock densities can be
responsible for high cannibalism and as a consequence reduce the larval stock
substantially. Tanasichuk (1998) believes that good recruitment is not directly
triggered by warm water, but is caused by strong upwelling in the area, preceding
the warm water events This idea would be in conformity with Smiles & Pearcy
(1971), who argue that off Oregon and California both spawning and recruitment
success depend strongly on upwelling conditions and related primary production.
The upwelling period would be the critical time for larvae to get enough phytoplankton for a better growth rate. Small changes in growth rates of krill larvae may
cause high fluctuations in recruitment (Daly 1990). A combination of predation,
food limitation and possibly cannibalism for eggs and larvae may restrict the period
of successful recruitment for E. pacifica.
In northern Japanese waters recruitment takes place after two months at 8 mm
length (Nishikawa et al., 1995). Japan, like California, is at the southern distribution
limit of this species, but at its southwestern distribution limit E. pacifica produces
only one generation every year (Iguchi et al., 1993). The conditions may be less
favourable in these waters than in the productive upwelling areas along the north
American coast. Small changes in the influence of the cold and productive Oyashio
current, versus the warm less productive Kuroshio current, negatively affect the E.
pacifica stock (Nishikawa et al., 1995; Taki & Ogishima, 1997; Yamamura et al.,
1998). However, the analysis of quantitative aspects of E. pacifica recruitment in
Japanese waters has still to be performed.
In the Barents Sea a long-term 12±14 year periodicity of overall krill density was
reported to be possibly coupled with a high-frequency 2±3 year variation in krill
recruitment (Boytsov & Drobysheva, 1987). Recruitment was defined as the transition from the larval into the juvenile age group 0+ and changes in recruitment were
explained by influences in water temperature on the youngest age group, so that the
Population Parameters
137
effects can be seen with a time lag of one year (Drobysheva, 1987). This would allow
predictions about possible recruitment success of T. inermis stocks.
Recruitment of E. superba was first defined as the proportion of recruits known as
the R1 rate which is the ratio of numbers in age class 1 to the numbers in age class 1
and all age classes above (de la Mare, 1994a). This procedure can also be applied for
age class 2. Since krill distribution is extremely patchy, neither stock density estimates nor density for each length class of standardised length±frequency distributions show a Gaussian distribution. De la Mare (1994b) analysed the underlying
statistics of these distributions and developed a new approach to create density-atlength (instead of length±frequency) data from scientific surveys as a basis for
recruitment estimates. However, some of the survey results on R1 and R2 given by
de la Mare (1994b) for the Atlantic sector are not representative, because these
calculations did not consider the incomplete survey coverage of the stock and its
underlying geographical non-random distribution pattern of different age groups
(Siegel et al., 1997).
One disadvantage of using the proportion of recruits is the possible misinterpretation of the values; a high proportionate value has different meanings when
stock density is either very high or very low. By standardising the number of recruits
to a given density (number of recruits/1000 m3), this new index was defined as
absolute recruitment of one-year-olds (RI1, (Siegel et al., 1998)).
Another disadvantage of the proportional index is that it does not necessarily
provide an indicator of reproductive success. This is because it does not refer to the
reproductive stock of the previous year, but to the year of the survey when
recruitment takes place, and the assumption of a closed krill population considered
to be in equilibrium is far from being realistic. In reality large variations in stock and
spawning stock size are quite apparent from krill net surveys, although these
changes in abundance usually occur gradually over several years and not from one
year to the next (Siegel et al., 1997, 1998). However, some conclusions can still be
drawn from the presently defined recruitment indices. The longest time series exists
for the Antarctic Peninsula region where proportional recruitment peaks were
observed for the year classes 1980±81, 1981±82, 1987±88, 1994±95. It is also evident
that most values from scientific surveys since the mid-1980s show generally poor
recruitment results which might explain part of the low stock size in the 1990s
(Siegel et al., 1997, 1998). No stock-recruitment relationship could be found for
Antarctic krill (Siegel & Loeb, 1995) and the very strong 1994±95 year class
demonstrated that even at a low spawning stock level strong cohorts can be produced.
Estimates of recruitment by the proportional method require unbiased sampling
regimes. Hence they are not strictly applicable to data derived from fisheries
operations. Given the low density of sampling for most areas, however, fisheriesderived indices can provide information that can augment scientific studies. Results
for the R2 values derived from Japanese Antarctic krill fishery data from the South
Shetland Island region show significant correlation with the R1 and R2 recruitment
138
Krill: Biology, Ecology and Fisheries
indices derived from scientific surveys (Kawaguchi et al., 1997). In contrast R1
values derived from the fishery were not significantly correlated with the scientific
data, probably due to net selectivity of the commercial trawl and/or incomplete
coverage of the areas where 1+ age-class krill tend to occur.
Immediately after the first recruitment estimates had been made, it became
obvious that the results differed substantially between years and regions. Various
parameters had been tested to find possible links between krill recruitment and the
biological and physical environment. The most promising variable to explain the
high variability in krill recruitment was sea-ice (Kawaguchi & Sarake, 1994; Siegel
& Loeb, 1995), especially the spatial extent and duration of winter sea-ice cover.
Since sea-ice seems to affect different krill population dynamics at a number of
scales, the following conceptual model was developed by Siegel & Loeb (1995)
derived from significant correlations of various parameters.
Extended sea-ice cover and a large spatial extent during winter is favourable for
an early onset of the spawning season, probably due to the ice-algae resource
available to krill during late winter to early spring. Protection from predation may
be another reason for a higher survival rate. As a consequence early reproduction
leads to a successful spawning event in summer. Larvae that hatch early in summer
show a high rate of survival during the feeding and growth season and are well
conditioned before their first winter. At the same time, salps which are thought to
compete for phytoplankton are scarce following winters of extensive ice cover. A
severe winter also favours the survival of the krill larvae under the ice, so that at the
end of the larval phase when entering the juvenile stage as one-year-olds (recruits)
we observe a high recruitment rate for that cohort.
In case of poor winter sea-ice conditions, salps as opportunistic short-living
animals develop rapidly in early spring and compete with krill for phytoplankton
(Loeb et al., 1997). With efficient competitors in the system and no additional seaice algae resource, krill may not be able to fulfil its energy requirement for growth
and an early spawning. Delayed or prolonged spawning leads to further competition
for newly hatched larvae with salps, and larvae show slower growth and therefore
according to Daly (1990) lower survival rates. An unsuccessful spawning event will
consequently be followed by poor recruitment in the next spring even if the following winter sea-ice conditions are favourable for larval survival.
Geographical differences in krill recruitment are quite obvious from Table 5.7. As
an example year classes 1980±81 and 1985±86 were highly successful in the Atlantic
sector, while recruitment in the Indian Ocean was a failure during the same years.
On the other hand year class 1988±89 experienced high recruitment in the Indian
Ocean, but it was low in the Atlantic. This is no surprise, because individual regions
of the Southern Ocean show significant interannual variability in ice coverage with
no coherence between the Antarctic Peninsula region and the Indian Ocean sector
(Stammerjohn & Smith, 1996). Due to these differences in the periodicity of ice
conditions for different sub-regions of the Southern Ocean, krill recruitment indices
have to be estimated and interpreted separately.
Populations of euphausiids are likely to suffer the impacts of long-term changes
1975/76
1976/77
1977/78
1978/79
1979/80
1980/81
1981/82
1982/83
1983/84
1984/85
1985/86
1986/87
1987/88
1988/89
1989/90
1990/91
1991/92
1992/93
1993/94
1994/95
1995/96
1996/97
1997/98
1998/99
Year
0.064
0.314
0.025
0.016
0.528
0.167
0.001
0.02
0.556
0.231
0.557
0.431
0.096
0.561
R1
R2
de la Mare, 1994b
Indian Ocean
0.156
0.651
0.057
0.099
0.375
0.000
0.068
0.046
0.622
0.198
0.120
0.559
0.757
0.470
0.030
0.0001
0.175
0.048
0.633
0.291
0.275
0.063
0.345
0.587
0.012
0.029
0.125
0.837
0.384
0.663
0.0001
0.214
0.069
0.144
Proportional recruitment (R)
R1
R2
Siegel et al., 1997, 1998
Antarctic Peninsula
0.574
0.719
0.000
0.058
0.336
0.000
0.462
0.585
0.156
0.147
0.191
0.316
0.194
0.000
0.000
0.427
0.142
0.072
0.165
0.334
0.360
0.000
0.000
0.083
0.129
0.169
0.000
0.000
0.451
0.086
R2
Kawaguchi et al., 1997
Livingston I.
Elephant I.
Table 5.7 Recruitment of Antarctic krill, Euphausia superba; year gives the time when the year-class was born.
4.31
51.98
0.88
0.64
7.45
0.00
2.24
3.37
74.62
42.07
3.06
90.21
245.93
130.05
4.59
0.01
16.71
Absolute R
RI1
N/1000 m3
Antarctic Pen.
Population Parameters
139
140
Krill: Biology, Ecology and Fisheries
caused by global warming or ozone depletion. Increasing air temperature will
affect sea-surface temperature and ice conditions in the Southern Ocean. Consequently this will affect krill recruitment and krill stock size in the long term.
Increased UV-B is another variable that might damage near surface krill concentrations, especially larvae, and increase mortality rates and again affect
recruitment rates and overall krill biomass (Jarman et al., 1999; Newman et al.,
1999). Current low recruitment rates over many years have to be regarded with
some concern. The krill stocks and the population dynamic parameters have to
be monitored carefully to gain a better understanding of the development of
these stocks as key elements of the ecosystems. The key question is have we been
observing high interannual variability over the past decade or is this the starting
point of a downward trend?
5.7
Conclusions
A large amount of effort has been devoted to the study of krill, particularly Antarctic krill, over the last 20 years. This has resulted in a number of developments in
methodology and in knowledge of certain key areas of their life history parameters.
Examples of these developments are the techniques for maintaining krill in
laboratories, the statistical methods developed for recruitment analysis and the
improvement in the tools available for length±frequency analysis. Despite these
advances there are still a number of basic aspects of the life history of these
organisms that require detailed study if we are to advance our understanding of krill
population dynamics. Specifically, improved, rapid techniques for ageing krill would
be extremely useful, validation of existing growth rate methodologies is necessary
and experimental studies into population parameters of species other than E.
superba are desirable. Additionally, further studies into the recruitment and mortality of a range of species of krill are required. Generally, there is still a major
problem in reconciling the results of field studies with those from the laboratory and
this should be a topic for future research. Many of the outstanding problems in krill
research could be addressed through concerted short-term (< 1 month) studies on
identified populations (or swarms) of krill which can be sampled repeatedly to
provide a time series of demographic measurements from an identified population.
The resulting information might then be compared to experimental results on
individuals obtained from the same population.
References
Aseev, Y.P. (1984) Size structure of krill populations and life span in the Indian Ocean Sector
of the Antarctic. Hydrobiology 6, 89±94.
Asthorsson, O.S. (1990) Ecology of the euphausiids Thysanoessa raschii, T. inermis and
Meganyctiphanes norvegica in Isafjord-Deep, northwest-Iceland. Mar. Biol. 107, 147±57.
Baker, A. de C. (1959) The distribution and life history of Euphausia triacantha Holt and
Tattersall. Discovery Rep. 29, 309±340.
Population Parameters
141
Baker, A. de C. (1963) The problem of keeping planktonic animals alive in the laboratory. J.
Mar. Biol. Ass. UK 43, 291±94.
Bamstedt, U. (1976) Studies on the deep-water pelagic community of Korsfjorden, Western
Norway; changes in the size and biochemical composition of Meganyctiphanes norvegica
(Euphausiacea) in relation to its life cycle. Sarsia 61, 15±30.
Basson, M. (1994) Towards a distribution of M/K for krill (Euphausia superba) required for
the stochastic krill yield model. CCAMLR-WG-Krill. 94/11, 16.
Basson, M. & Beddington, J. R. (1989) In University Research in Antarctica. Proceedings of
British Antarctic Survey Antarctic Special Topic Award Scheme Symposium. (R. B. Heywood, ed.), pp. 51±55. British Antarctic Survey, Cambridge.
Berkes, F. (1976) Ecology of Euphausiids in the Gulf of St. Lawrence. Journal of the Fisheries
Research Board of Canada 33, 1894±1905.
Beverton, R.J.H. & Holt, S.J. (1959) A review of the lifespans and mortality rates of fish in
nature, and their relation to growth and other physiological factors. CIBA Foundation
Colloquium on Aging 5, 142±77.
Bhattacharya, C. (1967) A simple method of resolution of a distribution into Gaussian
components. Biometrics 23, 115±35.
Bollens, S.M., Frost B.W. & Lin, T.S. (1992) Recruitment, growth, and diel vertical migration
of Euphausia pacifica in a temperate fjord. Mar. Biol. 114, 219±28.
Boysen, E. & Buchholz, F. (1984) Meganyctiphanes norvegica in the Kattegat ± studies of the
annual development of a pelagic population. Mar. Biol. 79, 195±207.
Boytsov, V.D. & Drobysheva, S.S. (1987) Effect of hydrometeorological factors on the
regularity of the long-term variations in euphausiid (Crustacea, Euphausiacea) abundance
in the southern Barents Sea. (Anonymous). In Proceedings of the third Soviet-Norwegian
Symposium, Murmansk, 26±28 May 1986. The effect of oceanographic conditions on distribution and population dynamics of commercial fish stocks in the Barents Sea. Institute of
Marine Research, Bergen, pp. 91±100.
Brattelid T.E. & Mathews J.B.L. (1978) Studies on the deep-water pelagic community of
Korsfjorden, western Norway. The dry weight and calorie content of Euchaeta norvegica
(Copepoda), Boreomysis arctica (Mysidacea), and Meganyctiphanes norvegica (Euphausiacea). Sarsia 63, 203±11.
Brinton, E. (1976) Population biology of Euphausia pacifica off southern California. Fish.
Bull. 74, 733±62.
Brinton, E. & Townsend, A.W. (1984) Regional relationships between development and
growth in larvae of Antarctic krill, Euphausia superba from field samples. J Crust. Biol. 4
(Spec No 1), 224±46.
Brodeur, R. D. & Pearcy, W. G. (1992) Effects of environmental variability on
trophic interactions and food web structure in a pelagic upwelling ecosystem.
Marine Ecology Progress Series 84, 101±19.
Buchholz, F. (1985) Moult and growth in euphausiids. (W.R. Siegfried, P. Condy & R.M.
Laws, eds), Antarctic nutrient cycles and food webs. In Proceedings of the 4th Symposium
on Antarctic Biology, Springer, Berlin. pp. 339±45.
Buchholz, F. (1991) Moult cycle and growth of Antarctic krill Euphausia superba in the
laboratory. Mar. Ecol. Prog. Ser. 69, 217±29.
Buchholz, F., Morris D.J. & Watkins, J.L. (1989) Analysis of field moult data: prediction of
intermoult period and assessment of seasonal growth in Antarctic krill Euphausia superba
Dana. Antarctic Sci. 1, 310±316.
142
Krill: Biology, Ecology and Fisheries
Clarke, A.C. (1976) Some observations on Krill (Euphausia superba Dana) maintained alive
in the laboratory. Brit. Ant. Surv. Bull. 43, 111±18.
Clarke, A. & Morris, D.J. (1983) Towards an energy budget for krill: the physiology and
biochemistry of Euphausia superba Dana. Polar Biol. 2, 69±86.
Clarke, A. & Peck, L.S. (1991) The physiology of polar marine zooplankton. Polar Res. 10,
355±69.
Constable, A.J. & de la Mare, W.K. (1996) A generalised model for evaluating yield and the
long-term status of fish stocks under conditions of uncertainty. CCAMLR Sci. 3, 31±54.
Dalpadado, P. & Ikeda, T. (1989) Some observations on moulting, growth and maturation of
krill (Thysanoessa inermis) from the Barents Sea. J. Plankton Res. 11, 133±39.
Dalpadado, P. & Skjoldal, H.R. (1991) Distribution and life history of krill from the Barents
Sea. Polar Res. 10, 443±60.
Dalpadado, P. & Skjoldal, H.R. (1995) Distribution and life cycle of krill north of 738N in the
Barents Sea, 1984±1992. Fisken og Havet 16, 1±50.
Dalpadado, P. & Skjoldal, H.R. (1996) Abundance, maturity and growth of the krill species
Thysanoessa inermis and T. longicaudata in the Barents Sea. Mar. Ecol. Progr. Ser. 144,
175±83.
Daly, K.L. (1990) Overwintering development, growth and feeding of larval Euphausia
superba in the Antarctic marginal ice edge zone. Limnol. Oceanogr. 35, 1564±76.
Daly, K.L. & Macaulay, M.C. (1988) Abundance and distribution of krill in the ice edge zone
of the Weddell Sea, austral spring 1983. Deep-Sea Res. 35, 21±41.
de la Mare, W.K. (1994a) Estimating krill recruitment and its variability. CCAMLR Sci. 1,
55±69.
de la Mare, W.K. (1994b) Modelling krill recruitment. CCAMLR Sci. 1, 49±54.
Drobysheva, S.S. (1987) Populational characteristics of abundant Barents Sea Euphausiacea.
ICES C.M. 12, 1±15.
Einarsson, H. (1945) Euphausiacea. 1. North Atlantic Species. Dana Rep. 27, 1±185.
Endo, Y. (1981) Ecological studies on the Euphausiids occurring in the Sanriku Waters with
special reference to their life history and aggregated distribution., Tohoku University,
Sendai, 166 pp.
Ettershank, G. (1983) Age structure and cyclical annual change in the Antarctic krill,
Euphausia superba Dana. Polar Biol. 2, 189±93.
Ettershank, G. (1985) Population age structure in males and juveniles of the Antarctic krill,
Euphausia superba Dana. Polar Biology 4, 199±201.
Everson, I. & Bone, D.G. (1986) Effectiveness of the RMT8 system for sampling krill
(Euphausia superba) swarms. Polar Biol. 6, 83±90.
Falk-Petersen, S. (1985) Growth of the Euphausiids Thysanoessa inermis, Thysanoessa
raschii, and Meganyctiphanes norvegica in a subarctic fjord, North Norway. Can. J. Fish.
Aquat. Sci. 42, 14±22.
Falk-Petersen, S., Gatten, R.R., Sargent, J.R. & Hopkins, C.C.E. (1981) Ecological investigations on the zooplankton of Northern Norway: seasonal changes in the lipid class composition of Meganyctiphanes norvegica, Thysanoessa inermis and Thysanoessa raschii. J.
Exp. Mar.Biol. Ecol. 54, 209±224.
Falk-Petersen, S. & Hopkins, C.C.E. (1981) Ecological investigations of the zooplankton
community of Balsfjorden, Northern Norway: population dynamics of the euphausiids
Thysanoessa inermis, Thysanoessa raschii and Meganyctiphanes norvegica in 1976 and
1977. J Plank Res 3, 177±92.
Farber-Lorda, J. (1990) Somatic length relationships and ontogenetic morphometric differ-
Population Parameters
143
entiation of Euphausia superba and Thysanoessa macrura of the southwest Indian Ocean
during summer (February 1981). Deep-Sea Res. 37, 1135±43.
Farber-Lorda, J. (1991) Multivariate approach to the morphological and biochemical differentiation of Antarctic krill (Euphausia superba and Thysanoessa macrura). Deep-Sea
Res. 38, 771±79.
Farber-Lorda, J. (1994) Length±weight relationships and coefficient of condition of
Euphausia superba and Thysanoessa macrura (Crustacea: Euphausiacea) in southwest
Indian Ocean during summer. Mar. Biol. 118, 645±50.
Fulton, J. & Le Brasseur, R. (1984) Euphausiids of the continental shelf and slope of the
Pacific coast of Canada. La Mer 22, 182±90.
Gomez, J.G. (1995) Distribution patterns, abundance and population dynamics of euphausiids Nyctiphanes simplex and Euphausia eximia off the west coast of Baja California,
Mexico. Mar. Ecol. Prog. Ser. 119, 63±76.
Gopalkrishnan, K. (1973) Development and growth studies of the euphausiid Nematoscelis
difficilis (Crustacea) based on rearing. Bull. Scripps Inst. Oceanogr. 20, 1±87.
Gros, P. & Cochard, J. (1978) Biologie de Nyctiphanes couchii (Crustacea, Euphausiacea)
dans le secteur nord du golfe de Gascogne. Ann. Inst. Oceanogr. 54, 25±46.
Hammer, C. & Braum, E. (1988) Quantification of age pigments (lipofuscin). Comp. Biochem. Physiol. 90, 7±17.
Harding, J.P. (1949) The use of probability paper for the graphical analysis of polymodal
frequency distributions. J. Mar. Biol. Ass. UK 28, 141±53.
Head, E., ed. (1997) Proceedings of the workshop on ecosystem considerations for krill and
other forage fisheries. Canadian Stock Assessment Proceedings Series, Fisheries and
Oceans Science, Canada, Halifax.
Hill, K.T. & Radkte, R.L. (1988) Gerontological studies of the damselfish, Dascyllus abisella.
Bull. Mar. Sci. 42, 424±34.
Hirche, H.-J. & Anger, K. (1987) The accumulation of age pigments during larval development of the spider crab, Hyas arenaeus (Decapoda, Majidae). Comp. Biochem. Phys. 88B,
777±82.
Hollingshead, K.W. & Corey, S. (1974) Aspects of the life history of Meganyctiphanes norvegica
(M. Sars) Crustacea (Euphausiacea) in Passamaquoddy Bay. Can. J. Zool. 52, 495±505.
Hopkins, C.C.E., Tande, K.S. Gronvik S. & Sargent, J.R. (1984) Biological investigations of
the zooplankton community of Balsfjorden, northern Norway: an analysis of growth and
overwintering tactics in relation to niche and environment in Metridia longa (Lubbock),
Calanus finmarchicus (Gunnerus), Thysanoessa inermis (Kroyer) and T. raschii (M. Sars).
J. exp. Mar. Biol. Ecol. 82, 77±99.
Hopkins, T.L. & Torres, J.J. (1989) Midwater food web in the vicinity of a marginal ice zone
in the western Weddell Sea. Deep-Sea Res. 36, 565±69.
Hosie, G.W. & Ritz, D.A. (1989). Body shrinkage in the sub-tropical euphausiid Nyctiphanes
australis G. O. Sars. J. Plank. Res. 11, 595±98.
Hosie, G.W. & Stolp, M. (1989) Krill and zooplankton in the western Prydz Bay region,
September±November 1985. Polar Biology 2, 34±45.
Hosie, G.W., Ikeda, T. & Stolp, M. (1988) Distribution, abundance and population structure
of the Antarctic krill (Euphausia superba Dana) in the Prydz Bay region, Antarctica. Polar
Biol. 8, 213±24.
Huntley, M. & Brinton, E. (1991) Mesoscale variation in growth and early development of
Euphausia superba Dana in the western Bransfield Strait region. Deep-Sea Research 38,
1213±40.
144
Krill: Biology, Ecology and Fisheries
Huntley, M.E., Nordhausen, W. & Lopez, M.D.G. (1994) Elemental composition, metabolic
activity and growth of Antarctic krill Euphausia superba, during winter. Mar. Ecol. Prog.
Ser. 107, 23±40.
Iguchi, N. & Ikeda, T. (1995) Growth, metabolism and growth efficiency of a euphausiid
crustacean Euphausia pacifica in the southern Japan Sea, as influenced by temperature. J.
Plank. Res. 17, 1757±69.
Iguchi, N., Ikeda, T. & Imamura, A. (1993) Growth and life cycle of a euphausiid crustacean
(Euphausia pacifica Hansen) in Toyama Bay, southern Japan Sea. Bull. Japan Sea Nat.
Fish. Res. Inst. 43, 69±81.
Ikeda, T. (1984) Development of the larvae of the Antarctic krill (Euphausia superba Dana)
observed in the laboratory. J. Exp. Mar. Biol. Ecol. 75, 107±117.
Ikeda, T. (1986) Preliminary observations on the development of the larvae of Euphausia
crystallorophias Holt and Tattersall in the laboratory. Memoirs of the National Institute of
Polar Research, Special Issue 40, 183±86.
Ikeda, T.P.G.T. (1987a) Longevity of the Antarctic krill (Euphausia superba Dana) based on
a laboratory experiment. Proc. Natl. Inst. Polar Res. Symp. Polar Biol. 1, 56±62.
Ikeda, T. (1987b) Mature Antarctic krill (Euphausia superba Dana) grown from eggs in the
laboratory. J. Plank. Res. 9, 565±69.
Ikeda, T. & Dixon, P. (1982) Body shrinkage as a possible over-wintering mechanism of the
Antarctic krill, Euphausia superba Dana. J. Exp. Mar. Biol. Ecol. 62, 143±51.
Ikeda, T. & Thomas, P.G. (1987a) Longevity of the Antarctic krill (Euphausia superba Dana)
based on a laboratory experiment. Proceedings of the NIPR Symposium on Polar Biology
1, 56±62.
Ikeda, T. & Thomas, P.G. (1987b). Moulting interval and growth of juvenile Antarctic krill fed different concentrations of the diatom Phaenodactylum tricornutum
in the laboratory. Polar Biology 7, 339±343.
Ikeda, T., Dixon, P. & Kirkwood, J.M. (1985) Laboratory observations of moulting, growth
and maturation in Antarctic krill (Euphausia superba Dana). Polar Biol. 4, 1±8.
Ikeda, T., Mitchell, A.W., Carlton, J.H. & Dixon, P. (1980) Transport of living Antarctic
zooplankton to a tropical laboratory: a feasibility study. Aust. J. Mar. Freshw. Res. 31, 271±74.
Ivanov, B.G. (1970) On the biology of the Antarctic krill (Euphausia superba Dana). Marine
Biology 7, 340±51.
Jarman, S., Elliott, N., Nicol, S., McMinn, A. & Newman, S. (1999) The base composition of
the krill genome and its susceptibility to damage by UV-B. Antarctic Sci. 11, 23±26.
Jarre-Teichmann, A. (1996) Initial estimates on krill. Fish. Centre Res. Rep. 4, 20.
Jillett, J.D. (1971) Zooplankton and hydrography of Hauraki Gulf, New Zealand. Mem. New
Zealand Oceanogr. Inst. 53, 1±103.
Jones, L.T., Forsyth, D.C.T. & Cooper, G.A. (1967) The occurrence of the two-spined form of
Thysanoessa inermis (Crustacea: Eupahsusiacea) in the North Sea. Bull. Mar. Ecol. 6, 181±84
Jorgensen, C. & Matthews, J.B.L. (1975) Ecological studies on the deep water pelagic
community of Korsfjorden, western Norway. Population dynamics of six species of
euphausiids in 1968 and 1969. Sarsia 59, 67±84.
Kanda, K., Takagi, K. & Seki, Y. (1982) Movement of the larger swarms of Antarctic krill
Euphausia superba off Enderby Land during 1976±77 season. J. Tokyo Univ. Fish. 68, 24±42.
Kawaguchi, S. & Sarake, M. (1994) Relationship between recruitment of the Antarctic krill
and the degree of ice cover near the South Shetland Islands. Fish. Sci. 60, 123±24.
Kawaguchi, K., Ishikawa, S. & Matsuda, O. (1986) The over wintering of Antarctic krill
Population Parameters
145
Euphausia superba Dana under the coastal fast ice off the Ongul Islands in Lutzow-Holm
Bay, Antarctica. Mem. Natl. Inst. Polar Res., Special Issue 44, 67±85.
Kawaguchi, S., Ichii, T. & Naganobu, M. (1997) Catch per unit effort and proportional
recruitment indices from Japanese krill fishery data in Subarea 48.1. CCAMLR Sci. 4, 47±
63.
Kawakami, T. & Doi, T. (1979) Natural mortality of krill and density in swarms. In Comprehensive report on the population of krill, Euphausia superba in the Antarctic. (ed Doi
T.) pp 19±21. Tokai regional Fisheries research Laboratory.
Komaki, Y. (1966) Technical notes on keeping euphausiids live in the laboratory, with a
review of experimental studies on euphausiids. Inf. Bull. Planktol. Japan 13, 95±105.
Kulka, D.W. & Corey, S. (1978) The life history of Thysanoessa inermis (Kroyer) in the Bay
of Fundy. Can. J. Zool. 56, 492±506.
Kulka, D.W. & Corey, S. (1982) Length and weight relationships of euphausiids and caloric
values of Meganyctiphanes norvegica (M. Sars) in the Bay of Fundy. J. Crust. Biol. 2, 239±47.
Labat, J. P. & Cuzin-Roudy, J. (1996) Population dynamics of the krill Meganyctiphanes
norvegica (M. Sars, 1857) (Crustacea: Euphausiacea) in the Ligurian Sea (NW Mediterranean Sea). Size structure, growth and mortality modelling. J. Plank. Res. 18, 2295±312.
Lancraft, T.M., Hopkins, T.L., Torres, J.J. & Donnelly, J. (1991) Oceanic micronektonic/
macrozooplanktonic community structure and feeding in ice covered Antarctic waters
during the winter (AMERIEZ 1988). Polar Biol. 11, 157±67.
Lasker, R. (1966) Feeding, growth, respiration and carbon utilisation of a euphausiid crustacean. J. Fish. Res. Board Can. 23, 1291±317.
Lasker, R.T. & Theilacker, G.H. (1965) Maintenance of euphausiid shrimps in the laboratory.
Limnol. Oceanogr. 10, 287±88.
Lindley, J.A. (1978) Population dynamics and production of euphausiids. I. Thysanoessa
longicaudata in the North Atlantic Ocean. Mar. Biol. 46, 121±30.
Lindley, J.A. (1980) Population dynamics and production of euphausiids. II. Thysanoessa
inermis and T. raschii in the North Sea and American coastal waters. Mar. Biol. 59, 225±33.
Lindley, J.A. (1980) Population dynamics and production of euphausiids. II. Thysanoessa
inermis and T. raschi in the North Sea and American coastal waters. Marine Biology 59,
225±33.
Lindley, J.A. (1982) Population dynamics and production of euphausiids. III. Meganyctiphanes norvegica and Nyctiphanes couchii in the North Atlantic and the North Sea.
Mar. Biol. 66, 37±46.
Lindley, J.A. & Williams, R. (1980) Plankton of the Fladen Ground during FLEX '76. II.
Population dynamics and production of Thysanoessa inermis (Crustacea: Euphausiacea).
Mar. Biol. 57, 79±86.
Loeb, V., Siegel, V., Holm-Hansen, O.R., Hewitt, R., Fraser, W., Trivelpiece, W. & Trivelpiece, S. (1997) Effects of sea-ice extent and salp or krill dominance on the Antarctic food
web. Nature 387, 897±900.
Lu, B. & Wang, R. (1996) Further study on distribution mixture analysis and its application to
age-groups analysis of Antarctic krill. Oceanologica et Limnologica Sinica 27, 179±86.
Macdonald, P.D.M. & Pitcher, T.J. (1979) Age-groups from size-frequency data: a versatile
and efficient method of analyzing distribution mixtures. J. Fish. Res. Board Can. 36, 987±
1001.
Mackintosh, N.A. (1967) Maintenance of living Euphausia superba and frequency of moults.
Norsk Hvalfangsttid 56, 97±102.
146
Krill: Biology, Ecology and Fisheries
Maklygin, L.G. & Latogursky, V.I. (1992) Possible approaches to the evaluation of the
Antarctic krill mortality. SC-CAMLR. WG-KRILL±92/8.
Marr, J.W.S. (1962) The natural history and geography of the Antarctic krill (Euphausia
superba Dana). Discovery Rep. 32, 33±464.
Marschall, H.-P. (1988) The overwintering strategy of Antarctic krill under the pack-ice of the
Weddell Sea. Polar Biol. 9, 129±35.
Matthews, J.B.L. (1973) Ecological studies on the deep-water pelagic community of Korsfjorden, western Norway. Population dynamics of Meganyctiphanes norvegica (Crustacea,
Euphausiacea) in (1968) and 1969. Sarsia 61, 15±30.
Mathews, J.B.L. & Hestad, L. (1977) Ecological studies on the deep-water pelagic communityof Korsfjorden, western Norway. Length/weight relationships for some macriplankton
organisms. Sarsia 63, 57±63.
Mauchline, J. (1960) The biology of the euphausiid crustacean Meganyctiphanes norvegica (M
Sars). Proc R Soc Edinb B (Biol). 67, 141±79.
Mauchline, J. (1966) The biology of Thysanoessa raschii (M Sars) with a comparison of its diet
with that of Meganyctiphanes norvegica (M Sars). In Some contemporary studies in Marine
Science (ed Barnes, H.). pp. 493±510. London: Allen and Unwin.
Mauchline, J. (1980) The biology of mysids and euphausiids. Adv. Mar. Biol. 18, 1±677.
Mauchline, J. & Fisher, L.R. (1969) The biology of euphausiids. Adv. Mar. Biol. 7, 1±454.
McClatchie, S. (1988) Food limited growth of Euphausia superba in Admiralty Bay, South
Shetland Islands, Antarctica. Cont. Shelf Res. 8, 329±45.
McClatchie, S., Rakusa-Suszczewski, S. & Filcek, K. (1991) Seasonal growth and mortality of
Euphausia superba in Admiralty Bay, South Shetland Islands, Antarctica. ICES J. Mar.
Res. 48, 335±42.
Miller, D.G. & Hampton, I. (1989) Biology and ecology of the Antarctic krill (Euphausia
superba Dana): a review. BIOMASS Scientific Series 9, 1±66.
Miller, D.G.M., Horstman, D.A., Wehmeyer, H.H. & Kuster, S. (1983) Maintenance of living
krill Euphausia superba in simple aquaria. S. Afr. J. Mar. Sci. 1, 65±69.
Morris, D.J., Watkins, J.L., Ricketts, C., Buchholz, F. & Priddle, J. (1988) An assessment of
the merits of length and weight measurements of Antarctic krill Euphausia superba. Br.
Antarctic Survey Bull. 79, 27±50.
Mullin, M.M. & Brooks, E.R. (1988) Extractable lipofuscin in larval marine fish. Fish. Bull.
US 86, 407±415.
Murano, M., Segawa, S. & Kato, M. (1979) Moult and growth of the Antarctic krill in the
laboratory. Trans. Tokyo Univ. Fish 3, 99±106.
Murphy, E.J., Watkins, J.L., Reid, K., Trathan, P.N., Everson, I., Croxall, J.P., Priddle, J.,
Brandon, M.A., Brierley, A.S. & Hofmann, E. (1998) Interannual variability of the South
Georgia marine ecosystem: biological and physical sources of variation in the abundance of
krill. Fish. Oceanogr. 7, 381±90.
Nemoto, T. (1957) Foods of baleen whales in the North Pacific. Sci. Repts. Whales Res. Inst.
Tokyo 12, 33±89.
Newman, S., Nicol, S., Ritz, D. & Marchant, H. (1999) Susceptibility of Antarctic krill
(Euphausia superba Dana) to ultraviolet radiation. Polar Biol. 22, 50±55.
Nicol, S. (1987) Some limitations on the use of the lipofuscin ageing technique. Mar. Biol. 93,
609±614.
Nicol, S. (1990) The age-old problem of krill longevity. BioSci. 40, 833±36.
Nicol, S. & Endo, Y. (1997) Krill Fisheries of the World, 100 pp. FAO, Rome.
Population Parameters
147
Nicol, S. & Stolp, M. (1991) Molting, feeding and fluoride concentration of the Antarctic krill,
Euphausia superba Dana. J. Crust. Biol. 11, 10±16.
Nicol, S., Kitchener, J., King, R., Hosie, G.W. & de la Mare, W.K. (2000) Population structure
and condition of Antarctic krill (Euphausia superba) off East Antarctica (80±1508E) during
the Austral summer of 1995/6. Deep-Sea Res. In press.
Nicol, S., Stolp, M. & Hosie, G.W. (1991) Accumulation of fluorescent age pigments in a
laboratory population of Antarctic krill (Euphausia superba Dana). J. Exp. Mar. Biol. Ecol.
146, 153±61.
Nicol, S., Stolp, M., Cochran, T., Geijsel, P. & Marshall, J. (1993) Growth and shrinkage of
Antarctic krill Euphausia superba from the Indian Ocean sector of the Southern Ocean
during summer. Mar. Ecol. Prog. Ser. 89, 175±81.
Nishikawa, J., Tsuda, A., Ishigaki, T. & Terazaki, M. (1995) Distribution of euphausiids in the
Kuroshio front and warm water tongue with special reference to the surface aggregation of
Euphausia pacifica. J. Plank. Res. 17, 611±29.
Pakhomov, E.A. (1995a) Demographic studies of Antarctic krill (Euphausia superba) in the
Cooperation and Cosmonaut Seas (Indian sector of the Southern Ocean). Mar. Ecol. Prog.
Ser. 119, 45±61.
Pakhomov, E.A. (1995b) Natural age-dependent mortality rates of Antarctic krill Euphausia
superba Dana in the Indian sector of the Southern Ocean. Polar Biol. 15, 69±71.
Pauly, D. (1980) On the interrelationships between natural mortality, growth parameters,
and mean environmental temperature in 175 fish stocks. J. Cons. Int. Explor. Mer. 39,
175±92.
Pilar, S.C. (1984) Laboratory studies on the larval growth and development of Euphausia
lucens (Euphausiacea). S. Afr. J. Mar. Sci. 2, 43±48.
Poleck, T.P & Denys, C.J. (1982) Effect of temperature on the molting, growth and
maturation of the Antarctic krill Euphausia superba (Crustacea: Euphausiacea) under
laboratory conditions. Marine Biology 70, 255±65.
Ponomareva, L.A. (1963) The euphausiids of the North Pacific, their distribution and
ecology. Dokl.Akad. Nauk. SSSR, 1±142.
Poulsen, E.M. (1926) Om den store lyskrebs (Meganyctiphanes norvegica) betydning som
fiskfode i Skagerrak. Dansk Fisk Tid. 24, 286±89.
Priddle, J., Croxall, J.P., Everson, I. et al. (1988) Large-scale fluctuations in distribution and
abundance of krill ± A discussion of possible causes. In Antarctic Ocean and Resources
Variability (ed Sahrage, D.) Berlin: Springer-Verlang.
Quetin, L. & Ross, R.M. (1991) Behavioural and physiological characteristics of the Antarctic
krill, Euphausia superba. Am. Zoologist 31, 49±63.
Quetin, L.B., Ross, R.M. & Clarke, A. (1994) Krill energetics: seasonal and environmental
aspects of the physiology of Euphausia superba. In Southern Ocean Ecology: the BIOMASS
perspective (S.Z. El-Sayed, ed.), pp. 165±84. Cambridge University Press, Cambridge.
Rosenberg, A.A., Beddington, J.R. & Basson, M. (1986) Growth and longevity of krill during
the first decade of pelagic whaling. Nature 324, 152±53.
Ross, R.M. (1981) Laboratory culture and development of Euphausia superba. Limnol.
Oceanogr 26, 235±46.
Ross, R.M. (1982) Energetics of Euphausia pacifica. I. Effects of body carbon and nitrogen
and temperature on measured and predicted production. Mar. Biol. 68, 1±13.
Ross, R. M. & Quetin, L. B. (1991) Ecological physiology of larval euphausiids, Euphausia
superba (Euphausiacea). Memoirs of the Queensland Museum 31, 321±33.
148
Krill: Biology, Ecology and Fisheries
Ross, R.M. et al. (1982) Energetics of Euphausia pacifica. II. Complete carbon and nitrogen
budgets at 88 and 128C throughout the life span. Mar. Biol. 68, 15±23.
Ross, R.M., Daly, K.L. & English, T.S. (1982) Reproductive cycle and fecundity of Euphausia
pacifica in Puget Sound, Washington. Limnol. Oceanogr. 27, 304±14.
Ross, R.M., Quetin, L.B., Vernet, M., Baker, K.S. & Smith, R.C. (2000) Growth limitation in
young Euphausia superba under field conditions. Limn. Oceanogr. in press.
Rudd, J.T. (1932) On the biology of the southern Euphausiidae. Hvalradets Skrifter 2, 1±105.
Sager, G. & Sammler, R. (1984) Seasonal length growth of the Norwegian krill (Meganyctiphanes norvegica) after data from Wiborg (1966±1969). Zoologische Jahrbucher/
Abteilung for Anatomie und Ontogenie 112, 79±84.
Sameoto, D.D. (1976) Respiration rates, energy budgets and moulting frequencies of three
species of euphausiids found in the Gulf of St. Lawrence. Journal of the Fisheries Research
Board of Canada 33, 2568±76.
Sheehy, M.J.R., Greenwood, J.G. & Fielder, D.R. (1994) More accurate chronological age
determination of crustaceans from field situations using the physiological age marker,
lipofuscin. Mar. Biol. 121, 237±45.
Siegel, V. (1982) Investigations on krill (Euphausia superba) in the Southern Weddell Sea.
Meeresforschung 29, 244±52.
Siegel, V. (1986) Structure and composition of the Antarctic krill stock in the Bransfield Strait
(Antarctic Peninsula) during the Second International BIOMASS Experiment (SIBEX).
Archiv fur Fischereiwiss 37, 51±72.
Siegel, V. (1987) Age and growth of Antarctic Euphausiacea (Crustacea) under natural
conditions. Mar. Biol. 96, 483±95.
Siegel, V. (1988) A concept of seasonal variation of krill (Euphausia superba) distribution and
abundance west of the Antarctic Peninsula. Antarctic Ocean and Resources Variability (D.
Sahrhage, ed.), pp. 219±230. Springer-Verlag, Berlin.
Siegel, V. (1989) Winter and spring distribution and status of the krill stock in the Antarctic
Peninsula waters. Archiv fur Fischereiwiss 39, 45±72.
Siegel, V. (1992a) Assessment of the krill (Euphausia superba) spawning stock off the Antarctic Peninsula. Archiv fur Fischereiwiss 41, 101±130.
Siegel, V. (1992b) Review of length±weight relationships for Antarctic krill, Selected Scientific Papers, 1992, SC-CAMLR-SSP/9, pp. 145±55. CCAMLR, Hobart, Australia.
Siegel, V. & Loeb, V. (1994) Length and age at maturity of Antarctic krill. Antarctic Sci. 6,
479±82.
Siegel, V. & Loeb, V. (1995) Recruitment of Antarctic krill Euphausia superba and possible
causes for its variability. Mar. Ecol. Prog. Ser. 123, 45±56.
Siegel, V., de la Mare, W.K. & Loeb, V. (1997) Long-term monitoring of krill recruitment and
abundance indices in the Elephant Island area (Antarctic Peninsula). CCAMLR Sci. 4, 19±
35.
Siegel, V., Loeb, V. & Groger, J. (1998) Krill (Euphausia superba) density, proportional and
absolute recruitment and biomass in the Elephant Island region (Antarctic Peninsula)
during the period 1977±1997. Polar Biol. 19: 393±98.
Smiles, M.C. & Pearcy, W.G. (1971) Size structure and growth rate of Euphausia pacifica off
the Oregon Coast. Fish. Bull. 69, 79±86.
Smith, S.L. (1991) Growth, development and distribution of the euphausiids Thysanoessa
raschii (M. Sars) and Thysanoessa inermis (Kroyer) in the southeastern Bering Sea. Polar
Res. 10, 461±78.
Population Parameters
149
Stammerjohn, S.E. & Smith, R.C. (1996) Spatial and temporal variability of Western Antarctic Peninsula sea ice coverage. (R.M. Ross, E.E. Hofmann, L.B. Quetin, eds), Ecological
Research West of the Antarctic Peninsula, 70, 81±104. American Geophysical Union,
Washington D.C.
Stretch, J.J., Hamner, P.P., Hamner, W.M., Michel, W.C., Cook, J. & Sullivan, C.W. (1988)
Foraging behaviour of Antarctic krill Euphausia superba on sea ice microalgae. Mar. Ecol.
Prog. Ser. 44, 131±39.
Sun, S. (1997) Using Antarctic krill as an indicator of environmental interannual change.
Korean J. Polar Res. 8, 97±103.
Sun, S. & Wang, R. (1995a) Ageing the Antarctic krill. Antarctic Res. 7, 59±62.
Sun, S. & Wang, R. (1995b) Study on the relationship between the crystalline cone number
and the growth of the Antarctic krill. Antarctic Res. 7, 1±6.
Sun, S. & Wang, R. (1996) Study on the relationship between the diameter of the compound
eye and the growth of the Antarctic krill. Antarctic Res. 7, 87±93.
Sun, S., de la Mare, W. & Nicol, S. (1995) The compound eye as an indicator of age and
shrinkage in Antarctic krill. Antarctic Sci. 7, 387±92.
Taki, K. & Ogishima, T. (1997) Distribution of some developmental stages and growth of
Euphausia pacifica Hansen in the northwestern Pacific on the basis of Norpac Net Samples.
Bull. Tohoku Nat. Fish. Res. Inst. 59, 95±117.
Tanasichuk, R.W. (1998) Interannual variations in the population biology and productivity of
Euphausia pacifica in Barkley Sound, Canada, with special reference to the 1992 and 1993
warm ocean years. Mar. Ecol. Prog. Ser. 173, 163±80.
Thomas, P.G. & Ikeda, T. (1987) Sexual regression, shrinkage, re-maturation and growth in
female Euphausia superba in the laboratory. Mar. Biol. 95, 357±63.
Vernet, M., Hunter, J.R. & Vetter, R.D. (1988) Accumulation of age pigments (lipofuscin) in
two cold-water fishes. Fish. Bull. 86, 401±407.
Vidal, J. & Smith, S.L. (1986) Biomass, growth and development of populations of herbivorous zooplankton in the southeastern Bering Sea during spring. Deep Sea Research 33,
523±56.
Wahle, R.A., Tully, O. & O'Donovan, V. (1996) Lipofuscin as an indicator of age in crustaceans: analysis of the pigment in the American lobster Homarus americanus. Mar. Ecol.
Prog. Ser. 138, 117±23.
Wang, R., Lu, B., Li, C., Wang, W. & Ji, P. (1995) Age-groups of Antarctic krill (Euphausia
superba Dana) by distribution mixture analysis from length-frequency data. Oceanologica
et Limnologica Sinica 26, 598±605.
Weigmann-Haass, R. (1997) Verbreitung von Makrozooplankton in der Gronlandsee im
Spatherbst 1988 (Crustacea: Ostracoda, Hyperiidea (Amphipoda), Euphausiacea). Helgolaender Meeresuntersuchungen 51, 69±82.
Wiborg, K.F. (1968) Atlantic euphausiids in the fjords of western Norway. Sarsia 33, 35±42.
Wiborg, K.F. (1971) Investigations on euphausiids in some fjords on the west coast of Norway
1966±1969. Fisk. Skr. Ser. Hav. 16, 10±33.
Yamamura, O., Inada, T. & Shimazake, K. (1998) Predation on Euphausia pacifica by
demersal fishes: predation impact and influence of physical variability. Mar. Biol. 132, 195±
208.
Zikov, L.A. & Slepokurov, V.A. (1982) The equation for the estimation of the natural
mortality of fishes (for example fish from Lake Endir) (in Russian). Rybn chozyaistvo 3,
36±37.
Chapter 6
Reproduction in Euphausiacea
Robin Ross and Langdon Quetin
6.1
Introduction
In the group Euphausiacea, females either retain mature eggs in a brood pouch until
they hatch, or release the eggs directly into the water column. All the species currently being commercially harvested, however, are broadcast spawners (Nicol &
Endo, 1997). The focus of our discussion of the reproduction patterns and strategies
of euphausiids is on those species that are of commercial interest, e.g. ThysanoÈessa
inermis, T. raschii, Meganyctiphanes norvegica, Euphausia pacifica, E. nana and E.
superba. However, we assume that certain reproductive characteristics are common
to the entire taxonomic group. Our understanding of the reproductive cycle will be
enhanced by including studies of non-harvested species, e.g. E. lucens, Nyctiphanes
australis, N. couchii and four additional Antarctic species, E. crystallorophias, T.
macrura, E. frigida and E. triacantha.
The emphasis in this chapter is on reproductive patterns and population fecundity
in euphausiids. Both are critical to our understanding of the pattern and level of
fishing effort a population can withstand over the long term. Reproductive patterns
include topics such as age and size at first spawning, initiation and duration of the
spawning season, and the distribution of spawning populations. OoÈcyte maturation
and the use of sexual development stages to identify seasonal patterns of ovarian
development are emphasised. We address also the topics of multiple spawning,
batch size, and embryo size. The implications of these reproductive characteristics
and strategies for fisheries management are briefly discussed.
6.2
Reproductive cycle
The patterns of reproduction in euphausiids show trends with latitude in timing and
duration of the spawning season. However, the exceptions yield clues about the
environmental conditions that support successful reproduction.
Patterns of reproduction ± age and size at first spawning
Euphausiids are relatively long-lived crustaceans, with life-spans that range from
slightly more than a year to seven years (Table 6.1). Smaller species and populations
150
±
33 or 35
25
11
22
21
11
±
12
Mat size
(mm)
2+
0.5
±
1+
0.5
2+ or 3+
1+
0.75±0.8
F 2+
M 1+
1
1st Spawn
(yr)
25
65
40
21
30
33
~18
12
25
Max size
(mm)
2+
5±7
2.5
~1
2+
4±5
1±1.2
<1
< 1.5 to > 2
Lifespan
(yr)
limited, 2 pks poss
limited
continuous
±
limited or
intermittent
limited, usually no pk
limited (long dur or 2 pks)
continuous, occ regression
limited
Reproductive
season
1±2
1±3
variable
8±10
1±3
1
8±10
±
1.5±2
Months
spawning
2
3±6
52
1
1
3±4
1
±
1
Years
spawning
Sources of information listed by species. E. c.: Harrington & Ikeda (1986); Siegel (1987). E. l: Pillar & Stuart (1988). E. n.: Nicol & Endo (1997). E. p.: Brinton (1976); Ross et
al., 1982; Nicol & Endo (1997). E. s.: Cuzin-Roudy (1987); Siegel (1987); Siegel & Loeb (1994); Nicol & Endo (1997). M. n: Nicol & Endo (1997). N. a.: Hosie & Ritz (1983);
Nicol & Endo (1997). T.i.: Nicol & Endo (1997). T. r.: Nicol & Endo (1997).
T. raschii
E. superba
M. norvegica
N. australis
T. inermis
E. crystallorophias
E. lucens
E. nana
E. pacifica
Species
Table 6.1 Reproductive patterns in euphausiids. Size and age at maturity, maximum size and lifespan, type of reproductive season (continuous,
intermittent, limited), number of months spawn during year, number of years an individual spawns; dur, duration; pk, peak.
Reproduction in Euphausiacea
151
152
Krill: Biology, Ecology and Fisheries
of a species living in lower latitudes, e.g. N. australis and E. pacifica, tend to have
life-spans between 1 and 1.5 years, whereas larger species and those in subarctic
regions live to about 2.5 years, e.g. M. norvegica, ThysanoÈessa spp. The two largest
polar euphausiids, E. crystallorophias and E. superba, have life-spans of over 4
years. Euphausiids usually reach maturity when they measure about half of their
maximum total length, and have reached 30 to 60% of their life-span (Table 6.1). In
several species of euphausiids, female maturity may be delayed beyond the point of
60% of the life-span. In T. inermis near Iceland, females mature at two years of age,
towards the end of their life-span, and a small percentage survive to breed in a
second reproductive season (Einarsson, 1945; Astthorsson, 1990). Siegel & Loeb
(1994) suggested that Antarctic krill, E. superba, shows `knife edge' maturity, i.e.
reaches maturity within a single age group. However, the high percentage of
immature females of this species in size classes > 40 mm in some summers suggests
that under some conditions reproduction may be postponed until the fourth summer
(Shaw, 1997).
Life-span and the age of first reproduction are not always the same for male
and female euphausiids. When there is a difference, males have a shorter lifespan. In the case of E. pacifica off Japan, females live 2+ to 3 years, whereas
males live less than 2 years (Nicol & Endo, 1997). A similar difference appears to
hold for E. crystallorophias in the Indian sector. Males live for 3+ years compared to the females' 4+ years (Pakhomov & Perissinotto, 1996). Although the
life-spans are different, the growth rates and age of first reproduction (2+) of
male and female E. crystallorophias are the same (Siegel, 1987). The implication
for both species is that males participate in one less reproductive season than
females. In other species males and females mature at different ages. Male T.
inermis near Iceland, for example, mature a year before the females, when they
are one year old not two (Astthorsson, 1990; Einarsson, 1945). However, T.
raschii males and females from the same area mature at the same age (Astthorsson, 1990).
A common assumption is that the energy costs of spermatophore production are
insignificant, and that female euphausiids expend more energy during the reproductive season than males. However, there are several lines of evidence that this
assumption does not hold for E. superba (Virtue et al., 1996), and thus may not hold
for other species. For E. superba, large mature males had higher mortality rates than
females, and lower total lipid and triacylglycerol stores. In addition, the sex ratio
decreased with increasing size, with the percentage of males decreasing sharply in
the size range 51±55 mm. Either the growth rates of males and females are significantly different, or mortality rates are different. In either case, the cost of
reproduction is one possible explanation for the differences observed. Little is
known about differences in male versus female growth or mortality rates for other
species, although Pillar & Stuart (1988) suggested that the retarded growth rates in
male E. lucens may be due to the allocation of energy to spermatophore production
throughout most of the year.
Reproduction in Euphausiacea
153
Reproductive patterns
Patterns of reproduction in euphausiids can be classified as continuous, intermittent
or limited (Table 6.1). With continuous reproduction there is no regression to a spent
stage once maturity is reached, although the intensity of spawning is higher during
some parts of the year. Continuous reproduction is primarily confined to species
inhabiting the mid-latitudes and equatorial regions. E. lucens is characterised as a
continuous spawner, with spawning most intense in late winter and early spring, just
prior to upwelling and bloom (Pillar & Stuart, 1988). In a few species, reproduction is
intermittent, with a cycle of maturation and regression more than once in one season,
resulting in multiple cohorts (generations) per year. Intermittent reproduction was
confirmed in N. australis, a species which produces three cohorts per year (Ritz &
Hosie, 1982), and E. pacifica in some regions (Brinton, 1976). Intermittent reproduction was inferred in two other species in the north-east Atlantic which produce
more than one generation per year, e.g. N. couchii, three to four per year (Lindley,
1982), and T. longicaudata, two per year (Lindley, 1978).
Limited reproduction means that there is a limited spawning season, with a cycle
of ovarian maturation and regression once a year. Egg production is limited to a 1.5
to 3 month season, alternating with a longer period of gonadal rest. Several peaks in
egg production can occur within this limited season, sometimes separated by periods
of weeks; such a pattern could lead to multiple cohorts in a single year class. Limited
reproduction is usually found in polar and subpolar species, e.g. the Antarctic
euphausiids, Thysanoessa spp., M. norvegica (Table 6.1). However, the pattern of
reproduction in a species is flexible. The reproductive season is limited over most of
the range of E. pacifica, a temperate and subarctic species, but intermittent in the
upwelling regions off Southern California (Brinton, 1976) and in some parts of
Puget Sound, Washington (Ross et al., 1982).
Timing and duration of spawning
From the early review of Mauchline & Fisher (1969), the timing and duration of
spawning were known to vary, both between populations of the same species living
in regions at similar latitudes, and among species at the same latitude (Fig. 6.1 and
Fig. 6.2). As a general rule, earlier and longer spawning occurs at lower latitudes.
These observations led to the hypothesis that timing and duration of spawning are
keyed to the period of elevated food production, i.e. earlier in the year and longer at
lower latitudes. In addition, spawning has been observed to coincide with the spring
bloom, e.g. T. inermis and T. longicaudata in the Barents Sea (Dalpadado &
Skjoldal, 1996), or times of high phytoplankton concentrations, e.g. E. pacifica in
areas of intermittent reproduction (Nicol & Endo, 1997).
The premise underlying the hypothesised link between food production and
spawning is that food must be available at the right time and in the right amounts to
satisfy the energetic demands of reproduction. Initiation of ovarian development
154
Krill: Biology, Ecology and Fisheries
Northern Hemisphere
Winter
Spring
Jan
Mar
Arctic
T. intermis
May
Bes
T. raschii
Jul
Fall
Sep
Nov
Bas
Bes
Bas
nNorw
Subarctic
T. inermis
BF
Jap
neAfl
T. raschii
Ice
Ice
Ak
M. norvegica
Norw
neAtl
JapS
Temperate
nwAtl
Clyde
E. pacifica
E. pacifica
Summer
PuS
nPac
BrC
Sanriku, Jap
SoCa
Or
M. norvegica Med
N. couchii n Atl
Fig. 6.1 Initiation, duration and intensity of spawning season for euphausiid species in different
regions in the Northern Hemisphere. BeS, Bering Sea; BaS, Barents Sea; nNorw, northern Norwegian fjords; Ice, Icelandic fjords; BF, Bay of Fundy, Canada; Jap, Japan; ne- or nw, northeast or
northwest: Atl, Atlantic Ocean; Ak, Auke Bay, Alaska; Clyde, Firth of Clyde, Scotland; nPac,
north Pacific Ocean; BrC, British Columbia; PuS, Puget Sound, Washington; JapS, Japan Sea; Or,
Oregon coast; SoCa, southern California coast; Med, Mediterranean Sea. Sources of information
listed by species. T. i.: Kulka & Corey (1978); Hanamura et al. (1989); Astthorsson (1990);
Einarsson (1945); Smith (1991); Nicol & Endo (1997). T. r.: Falk-Petersen & Hopkins (1981);
Astthorsson (1990); Paul et al. (1990); Einarsson (1945); Timofeyev (1994); Smith (1991); Nicol &
Endo (1997). M. n.: Lindley (1982); Boysen & Buchholz (1984); Nicol & Endo (1997); Cuzin-Roudy
(1993); Cuzin-Roudy & Buchholz (1999). E. p.: Ponomareva (1966); Smiles & Pearcy (1971);
Brinton (1976); Heath (1977); Ross et al. (1982); Bollens et al. (1992); Iguchi et al. (1993); Nicol &
Endo (1997). N. c.: Lindley (1982).
may be keyed to a seasonal cue such as photoperiod, but actual spawning (release of
eggs) will depend on the rate of ova maturation, and thus the rate of energy input
prior to spawning. In some species energy is stored from the previous season for
ovarian development, in others ovarian development is dependent on immediate
food supplies. If reproduction is iteroparous (Section Population fecundity), the rate
of production and number of subsequent batches of eggs is likely to depend on both
energy input prior to spawning, and on food availability during the spawning season.
Reproduction in Euphausiacea
155
Southern Hemisphere
Winter
Jul
Spring
Sep
Summer
Nov
Jan
Fall
Mar
May
Temperate
N. australis
E. lucens
Antarctic Circumpolar Current
E. triacantha
E. frigida
T. macrura
Sea Ice Zone
E. triacantha
E. frigida
T. macrura
E. superba
E. crystallorophias
WAP
WAP
WAP
Wed
RossS
WAP
PrdB
Wed
RossS
WAP
Wed
Fig. 6.2 Initiation, duration and intensity of spawning season for euphausiid species in different
regions in the Southern Hemisphere. Wed, Weddell gyre; WAP, western Antarctic Peninsula; RossS,
Ross Sea; PrdB, Prydz Bay and Lasarev Sea. Sources of information listed by species. N. a.: Blackburn
(1980); Ritz & Hosie (1982). E. l.: Pillar & Stuart (1988); Stuart & Pillar (1988). E. t.: Baker (1959).
E. f.: Makarov (1977); Siegel (1986); Makarov et al. (1990); Men'shenina (1992). T. m.: Stepnik (1982);
Makarov et al. (1990); Men'shenina (1988, 1992); Makarov & Men'shenina (1992). E. s.: Makarov et
al. (1985, 1990, 1991); Hosie & Cochran (1994); Makarov & Men'shenina (1992); Ross & Quetin
(1983); Quetin et al. (1994). E. c.: Fevolden (1980); Makarov et al. (1990, 1991); Brinton & Townsend
(1991); Makarov & Men'shenina (1992); Harrington & Thomas (1987).
The sequence of spawning for Antarctic euphausiids illustrates the latitudinal
trend. In general, production cycles in the northern areas start earlier, and finish
later, than further south (Hart, 1942; Foxton, 1956). Matching this cycle, oceanic
species at lower latitudes reproduce in the early spring, whereas the shelf and
coastal species at higher latitudes spawn one to two months later (Fig. 6.2). The
exception is the coastal species, E. crystallorophias, which reproduces before E.
superba (Fig. 6.2). Also, at higher latitudes where the period of elevated production
is longer, spawning seasons are longer, e.g. west of the Antarctic Peninsula compared to the Ross Sea, and the Scotia Sea compared to the Lasarev Sea. Spiridonov
(1995) confirmed this pattern for E. superba in a study of reproduction with data
from both the Weddell Sea and waters west of the Antarctic Peninsula. He identified five habitats based on the timing of initiation, duration and intensity of
156
Krill: Biology, Ecology and Fisheries
spawning. In the northern most habitat, at the fringe of the extent of seasonal seaice, spawning was early (late November±early December), and tended to be long
(3±3.5 month) but variable in duration. In contrast, in coastal areas at higher latitudes, spawning began later and was of shorter duration (~ 1.5 month).
Understanding this link between food production and spawning allows us to
reconcile what appear to be conflicting views of the influence of seasonal sea-ice
dynamics on the spawning season of E. superba (Spiridonov, 1995; Siegel & Loeb,
1995; Makarov & Men'shenina, 1992). For Antarctic krill living at the tip of the
Antarctic Peninsula, Siegel & Loeb (1995) found that low sea-ice extent in winter
and/or early sea-ice retreat led to a delay in spawning. In contrast, Spiridonov
(1995) concluded that low sea-ice extent resulted in early spawning, and a slow
(late) retreat resulted in a delayed spawning season that was not intensive. In coastal
waters in the Weddell Sea, when polynas open early in the spring, i.e. simulating an
early retreat, reproduction starts earlier than in waters to the north which are still
covered with ice. Makarov & Men'shenina (1992) called this phenomenon the
polyna or `oasis' effect, and it can reverse the normal north±south gradient in timing
of reproduction for both E. superba and T. macrura.
Can we reconcile these disparate views of the influence of seasonal sea-ice on the
timing and intensity of spawning of Antarctic krill? The retreat of sea-ice impacts
food availability in spring because the melting sea-ice conditions the water column
for ice edge blooms. Without this conditioning, phytoplankton blooms are delayed
until late spring or early summer. Since Antarctic krill do not come into reproductive condition under the ice (Cuzin-Roudy & Labat, 1992), the extent and timing
of these ice edge blooms will play an important role in the rate of ovarian
development and thus the timing of spawning. Sea-ice extent represents the areal
extent of conditioning and thus enhanced food availability in spring; the timing of
retreat represents when conditioning occurs and thus the timing of the blooms. The
timing of sea-ice retreat is region-specific, that is, `early' retreat for northern regions
is likely to be earlier in the spring than in southern regions. `Early' in one region may
lead to blooms prior to the peak demand for food for ovarian development, whereas
in another region `early' retreat may create ideal conditions. In the Elephant Island
area (Siegel & Loeb, 1995), at the northern edge of the influence of seasonal sea-ice,
if sea-ice in winter is low little of the region will be conditioned for an ice edge
bloom, and the area of enhanced food availability will be small. With an early
retreat, bloom conditions may occur prior to the time of maximum need for food for
ovarian development. Thus low winter sea-ice extent and/or early retreat mean
lower food availability in the spring, and these environmental conditions will be
associated with slow ovarian maturation, and delayed spawning. However, when
polynas open early in the spring in the higher latitude coastal waters of the Weddell
sea, ice edge blooms create a source of food at a time of prime need. Polynas allow
the cycle of food production to start earlier in the spring at higher latitudes than in
the ice-covered seas at lower latitudes; food is available for ovarian maturation, and
thus spawning can start earlier. The reverse will be found with a `late' retreat. The
Reproduction in Euphausiacea
157
two viewpoints can be reconciled by assuming that sea-ice mediates the areal extent
and timing of spring food for ovarian development.
The Antarctic species are not the only euphausiids to show inter-annual and
geographical variability in reproduction within a species, even among locations that
differ only in food availability and not in either temperature or photoperiod. The
reproductive season for the Northern krill, M. norvegica, begins in the spring in all
areas but the north-west Atlantic off Canada where the influence of the cold
Laborador Current delays the onset of net primary production, and spawning does
not begin until summer (Lindley, 1982). For the smaller E. pacifica, the population
in Puget Sound begins to spawn in the spring, continuing through summer in some
areas of Puget Sound but not others (Ross et al., 1982; Bollens et al., 1992). However,
E. pacifica off the coast of Oregon does not spawn until late summer and early fall
(Smiles & Pearcy, 1971) (Fig. 6.1). Peaks in chlorophyll a occur in April and May in
Puget Sound, but off Oregon appear later in the year and are associated with the
summer upwelling regime. A link between the phytoplankton bloom and the length
and intensity of spawning has also been suggested as the source of the inter-annual
differences in the duration of the period of high egg concentrations of T. raschii in
Auke Bay, Alaska (Paul et al., 1990).
Spawning is not usually synchronous among species where their ranges overlap.
In the arctic and subarctic North Atlantic, T. inermis spawns earlier than T. raschii,
which in turn spawns earlier than M. norvegica when that species is present
(Astthorsson, 1990). Such differences may reflect the extent to which reproduction
depends on that season's primary production, whether directly or indirectly, versus
stored reserves from the previous season. Omnivores or carnivores may show an
indirect dependence on primary production through food web links. If the populations of prey increase in response to seasonal phytoplankton blooms, food availability for the omnivores and carnivores is indirectly dependent on periods of
primary productivity. In some species, initiation of ovarian maturation may be
keyed so spawning occurs before the spring bloom or period of upwelling. With this
strategy, stored reserves provide the energy to initiate reproduction, and seasonal
production benefits the larvae. If reproduction is in response to the spring bloom or
period of upwelling, the implication is that stored reserves are not adequate to fuel
reproduction. The degree of inter-annual variability in the initiation of the spawning
season is thus apt to vary with the degree of reliance on stored reserves versus that
season's food production. In a recent comparison of the lipid biochemistry of the six
most abundant species of Arctic and Antarctic euphausiids, Falk-Petersen et al.
(August 1999) found that neutral lipid deposits are primarily accumulated for
reproduction. Four species, T. inermis, T. longicaudata, T. macrura and E. crystallorophias, depend primarily on lipid reserves for spring reproduction. T. raschii and
E. superba rely on food availability in the spring for final gonad maturation. E.
superba is the single Antarctic euphausiid with a reproduction season in summer,
and a recent study (Shaw, 1997) suggests that the initiation and duration of the
spawning season is dependent on both spring and summer food sources.
158
Krill: Biology, Ecology and Fisheries
Mating and spawning
In euphausiids, males tend to mature several months earlier in the season than
females. For example, in the northern Baltic, immature male M. norvegica are
distinguishable in October whereas immature females cannot be distinguished until
January; the males also have fully developed spermatophores at nine months
(Boysen & Buchholz, 1984). Mating behaviour begins up to 1±2 months prior to
spawning and continues throughout the spawning season (Table 6.2). Thus transfer
of the spermatophore occurs a considerable time before spawning, and one cannot
assume that the presence of spermatophores on a female indicates that the female is
gravid. Euphausiids lose the sperm mass held in the thelycum when they moult.
Since euphausiids moult on a time-scale of weeks throughout their lives, mating
must continue throughout the reproductive season to ensure that eggs are fertilised.
Males with mature spermatophores in the ejaculatory ducts are found throughout
the spawning season (Bargmann, 1937; Boysen & Buchholz, 1984; Astthorsson,
1990).
Table 6.2 Relative timing of mating behaviour and initiation of spawning.
Species
Mating begin
Spawning begin
Reference
E. superba
Nov
late Dec
Bargmann (1945)
M. norvegica
Jan
Apr±Oct
Falk-Petersen & Hopkins (1981)
Boysen & Buchholz (1984)
T. inermis
late Feb
mid±late Apr
Hanamura et al. (1989)
T. inermis
Feb
Apr
Astthorsson (1990)
T. raschii
Apr
May
Mating behaviour has been observed in E. superba in the ice edge zone in
December (Naito et al., 1986). Bargmann (1937) described the exchange of spermatophores in E. superba. A sequence of chase and contact behaviour between
male and female E. superba believed to be part of mating behaviour has also been
observed in large aquaria in the laboratory (Fig. 6.3) (Ross et al., 1987). Durations of
chases were about 5 sec, and the contact or close times 10±30 sec. In the tanks males
appeared to perceive a chemical trail within 8±10 cm of the female, and turn in
pursuit.
Surface aggregations have often been described as facilitating mating behaviour
(Ritz, 1994). These aggregations tend to occur in the late winter or early spring,
prior to reproduction. In five species, mature breeding individuals predominated in
daytime surface aggregations (Table 6.3). The aggregations thus contain mature
euphausiids in the close proximity necessary to initiate a pursuit sequence at a time
in the cycle when finding a mate is critical. The sex ratios in these surface swarms
were sometimes (Nemoto et al., 1981, E. superba; Nicol, 1984, M. norvegica), but not
Reproduction in Euphausiacea
159
Fig. 6.3 Relative positions of the male (on the bottom) and female adult Euphausia superba
during the chase and contact sequence of mating behaviours.
Table 6.3 Observations of daytime surface aggregations dominated by mature individuals. See
Fig. 6.1 for the key.
Species
Months, or
Location
Dom sex
or Size
Reference
E. pacifica
Mar±Apr, Japan
nw
female
> 14 mm
Endo (1984)
Odate (1991) cited in Nicol & Endo (1997)
May±June,
nPac
E. superba
Dec in ice zone
Kotori (1995)
mating
Naito et al. (1986)
Nicol et al. (1987)
O'Brien (1988)
M. norvegica
female
Nicol (1984)
T. spinifera
±
Smith & Adams (1988)
T. inermis
Feb±Apr
female
Hanamura et al. (1989)
T. raschii
Barents Sea
female
Timoteyev (1994)
always extreme (O'Brien, 1988, N. australis). In another study of aggregation
characteristics in E. superba, population composition was often skewed, with
mature males and gravid females dominating some aggregations, and immature
individuals dominating others (Watkins et al., 1992). However, there was a positive
correlation between the presence of mature males and mature females with
developing ovaries (Watkins et al., 1992). Such an association would be necessary
for the continuing transfer of spermatophores to females throughout the reproductive season. There also appears to be spatial segregation on small scales. Off
Japan, male and female E. pacifica may separate after mating. Males dominate the
surface swarms during the day, with females at depths below 15 m (Nicol & Endo,
160
Krill: Biology, Ecology and Fisheries
1997). Sex ratios in catches of M. norvegica also vary (Hollingshead & Corey, 1974),
with surface and deep aggregrations showing different population structures that
suggest that mated and ready-to-spawn females separate from the rest of the
population (Nicol, 1984; Cuzin-Roudy & Buchholz, 1999).
Although information is scarce, the distribution of spawning populations of
euphausiids does not appear to be homogeneous. The concept of spatial and geographical succession in Antarctic krill developmental stages has been described by
several authors (Makarov & Sysoyeva, 1985; Siegel et al., 1990; Lascara et al., 1999).
Gravid female E. superba are generally found in the vicinity of the continental slope
(Siegel, 1992), and in waters overlying Circum-Polar Deep Water (CDW) (Hofmann et al., 1992). Release of eggs over CDW is thought to be part of a reproductive
strategy to enhance survival of the larvae. During the season, as the females progressively mature, their distribution changes. Less advanced stages of female E.
superba are found on the inner shelf (Shaw, 1997) or closer to the ice edge (CuzinRoudy & Labat, 1992; Quetin et al., 1992). Females in vitellogenesis and spawning
are first found in open waters. Spawning in several other species appears to be
spatially restricted. Spawning zones for T. macrura are found in the Antarctic
Convergence, and the Secondary Frontal Zone (Men'shenina, 1988). There is a
single spawning centre for T. inermis in the waters south of Spitzbergen, with eggs
and larvae transported elsewhere (Timofeyev, 1994). And lastly, the percentage of
gravid female E. lucens increases offshore (Pillar & Stuart, 1988). Thus, in these four
cases, spawning is not randomly distributed within the species' range. The advantages to this heterogeneity are not apparent for all species, but the patterns are clear.
Such distributions may enhance retention of larvae in favourable regions as has
been found for fishes or enhance hatching success or reduce predation on the vulnerable early life history stages.
The concept of `pseudopopulations' or non-breeding populations should also be
mentioned in the context of the distribution of maturity stages of euphausiids.
Although most species breed throughout their range of occurrence, there are
exceptions. T. inermis, for example, does not breed north of 65±708N (Nicol &
Endo, 1997), thus the population of T. inermis in the southern Barents Sea
(Zelikman et al., 1980) is believed to be non-reproducing. There is also some debate
over whether populations of E. superba around South Georgia are non-reproducing.
Although abundances can be high, the lack of eggs and young larvae in these regions
is cited as evidence for populations that are not successfully reproducing.
Ovarian maturation cycle
Ovarian development and maturation occurs over a period of several months prior
to spawning. In T. inermis, ovaries mature over the three month period between
mid-January and mid-April, prior to spawning in the spring. Ovarian development
in T. raschii in the north-east Pacific (Zelikman, 1958), M. norvegica in the northeast Atlantic (Zelikman, 1958), and E. superba in the Antarctic (Bargmann, 1945)
Reproduction in Euphausiacea
161
also takes about three months. Maturation of testes in males generally is more rapid,
about 1 month in T. inermis (Kulka & Corey, 1978) and 2+ months in M. norvegica
(Mauchline, 1968; Hollingshead & Corey, 1974).
General progression
The details of ovarian maturation help us to understand where environmental
variability is most likely to impact rates of ovarian maturation. Ovarian maturation
in crustaceans is characterised by a progression in ova development that is found in
all species (Nelson, 1991). First, during oogenesis the ooÈgonia (og) multiply in the
germinal zones (GZ) of the ovary. When the ooÈgonia leave the germinal zone, they
are characterised as primary or young ooÈcytes (yoc). The yocs represent multiple
batches of eggs in the sense that groups of eggs will develop in sequence, not
simultaneously. At this point the female may enter a resting phase. Upon leaving the
resting phase, groups of yoc accumulate glycoproteic yolk synthesised within the
ooÈcyte in a process called primary or pre-vitellogenesis. At this point the ovary is
increasing in size and mating may occur, but the female euphausiid is not swollen.
Mating has occurred when either the thelycum is full or spermatophores are
attached. The growing type 1 ooÈcytes (oc1) may accumulate at a second plateau
before they enter vitellogenesis. During vitellogenesis, or lipidic yolk accumulation
in type 2 (oc2) and type 3 (oc3) ooÈcytes, the ooÈcytes rapidly enlarge, increasing the
cytoplasm to nucleus ratio. The source of the lipidic yolk is outside the ooÈcyte, likely
the `fat body' (Cuzin-Roudy & Buchholz, 1999). Final egg maturation occurs after
vitellogenesis is complete, the nucleus moves to the periphery and loses its membrane, and meiosis occurs in the type 4 ooÈcyte (oc4). Another plateau can occur at
this point. Ovulation, fertilisation and release of the embryos are the final steps.
The plateaus are assumed to occur at points during ovarian maturation when
additional resources or environmental cues such as photoperiod are needed for
further development. The cue for the second plateau is suggested to be a threshold
accumulation of `lipid' in the `fat body', enough to complete the development of one
group of ooÈcytes. Evidence for the existence of a `fat body' in euphausiids is
emerging for two species, E. superba and M. norvegica (Cuzin-Roudy & Buchholz,
1999).
Maturity keys based on physiological state
As ooÈcytes move through the succession of physiological steps of ovarian maturation as described above, they show characteristic morphological (Fig. 6.4) and
biochemical changes. Keys for stages of female sexual development for two
euphausiids, E. superba (Table 6.4, Cuzin-Roudy & Amsler, 1991) and M. norvegica
(Cuzin-Roudy, 1993) have been developed based on both external secondary sex
characteristics and the types of ooÈcytes in the ovary as characterised with the
`squash' technique. External secondary sex characteristics include the development
162
Krill: Biology, Ecology and Fisheries
Fig. 6.4 Relative size of ooÈcytes and nucleus/cytoplasm ratio in the cycle from gametogenesis to
final maturation. yoc, young ooÈcytes or primary ooÈcytes; oc1, stage 1 ooÈcytes; oc2, stage 2 ooÈcytes;
oc3, stage 3 ooÈcytes; oc4, stage 4 ooÈcytes. Nucleus is clear. Cytoplasm becomes increasingly
speckled as lipidic yolk accumulation proceeds.
of the thelycum, the size of the ovary as seen through the carapace, and the degree
of swelling of the thorax. During gametogenesis the thelycum is not yet present, and
the ovary is composed of germinal zones and ooÈgonia, small non-descript cells.
During oogenesis the thelycum is immature, and although the ovary is still small,
some of the ooÈgonia have transformed into the slightly larger yoc. During previtellogenesis the thelycum is completely developed. The ovary grows considerably
during this phase, but the thorax is not swollen. The ooÈcytes (oc1) accumulate
Reproduction in Euphausiacea
163
Table 6.4 Definitions of sexual development stages (SDS) for female Euphausia superba (CuzinRoudy & Amsler, 1991).
SDS
Dev phase
Thelycum
Ovary and ooÈcytes
1
Gametogenesis
Absent
Ovary small. Germinal zone
(GZ) and primary and
secondary ooÈgonia (og 1, og 2)
2
Oogenesis
Present, sternal and
coxal plates not fully
grown
Ovary small. Og 2 developed
into yoc
3
Previtellogenesis
Mature thelycum
Thorax not swollen. Oc1 up to
200 mm
4
Previtellogenesis
Mature thelycum, may
be red. May be mated
Thorax not swollen.
Developing ovary visible. Oc1
up to 450 mm
5
Vitellogenesis
Mature thelycum, red.
Mated
Ovary obvious. Oc2 present
6
Vitellogenesis
Mature thelycum, red.
Mated
Thorax conspicuously swollen.
Ovary full of oc3; may have
various combinations of GZ,
oc1, oc2 and oc3
7
OoÈcyte maturation
Mature thelycum, red.
Mated
Ovary in live krill opaque or
bluish, full of oc4; may have
various combinations of GZ,
oc1, oc2, and oc4
8
Oviposition
Mature thelycum, red
Thorax swollen, but ovary
reduced in size and lobes
contracted. May have various
combinations of GZ, oc1, oc2
and residual oc4 (continuing)
or only GZ and residual oc4
(final or post spawn)
9
Ovarian reorganization
Mature thelycum, red
Thorax not swollen. Ovarian
lobes irregular and lumpy,
contain GZ with og1, og2 and
yoc, plus residual oc4
10
Ovarian reorganization
Thelycum regressing to
more juvenile form
Thorax not swollen. Ovarian
lobes irregular and lumpy,
contain GZ with og1, og2 and
yoc, possibly residual oc4
glycoproteic yolk, and are translucent with a nucleus/cyctoplasm ratio of about 60%.
Mating occurs during previtellogenesis, and a sperm mass often fills the thelycum.
During vitellogenesis, the ooÈcyte grows, and the accumulation of lipidic yolk gives
the cytoplasm in the ooÈcyte a cloudy or grainy appearance (oc2s and oc3s). Opaque
ooÈcytes overlapping the size of oc1s but with a cloudy appearance are oc2s; larger
ooÈcytes with a lower nucleus/cytoplasm ratio and a more granular appearance are
164
Krill: Biology, Ecology and Fisheries
oc3s. The nucleus is large and distinct in both oc2s and oc3s. Externally the ovary is
now conspicuous, and the thorax swollen. Ovaries with mature ooÈcytes (oc4) appear
opaque or even bluish gray, e.g. E. crystallorophias (Harrington & Thomas, 1987),
E. pacifica (Ross et al., 1982), E. superba (Ross & Quetin, 1983), and M. norvegica
(Cuzin-Roudy & Buchholz, 1999). Oc4s have completed vitellogenesis, lack a large
central nucleus, and meiotic figures can be seen under high magnification. The small
nucleus without a membrane is in the peripheral cytoplasm (Fig. 6.4). After the
mature ooÈcytes (oc4s) are released, the thorax remains swollen, and the female is
considered post-spawn. However, the female may not have completed the reproductive season. In both E. superba (stage 8c = 8 continuing) and M. norvegica (stage
5 1) the presence of ooÈcytes in earlier stages indicates that another group or batch of
ooÈcytes will mature and another spawning will occur that season. When the ovary in
a post-spawn female contains only the germinal zone and oogonia, the reproductive
season is complete for that individual, and she enters a period of reorganisation and
reproductive rest.
The sexual development stages (SDS) for E. superba (Table 6.4) have been
slightly simplified for use with M. norvegica (Cuzin-Roudy, 1993). For M. norvegica, no stage is defined for gametogenesis, and there is only one stage each for
previtellogenesis and vitellogenesis. The distinction between post-spawn females
that will recycle and those that have completed egg production for that season is
retained (see comparison in Table 6.5). Similar keys could be used for other
species, since all euphausiids move through the same stages during ovarian
maturation.
Table 6.5 Sexual development stages of euphausiids and the relationship to known physiological
phases of ovarian development.
Physiological phases of ovarian
development
Euphausia superba
(Cuzin-Roudy &
Amsler, 1991)
Meganyctiphanes
norvegica
(Cuzin-Roudy, 1993)
Immature or resting
2 and 10
1
3 and 4
5 and 6
2
3
7
8c and 8f
4
51 and 52
9
6
Oogenesis
ONSET
OoÈcyte development
. previtellogenesis; mating activity
. vitellogenesis
OoÈcyte maturation and egg release
. ready to spawn
. post-spawn, may recycle
END
Spent females, ovary in regression
Reproduction in Euphausiacea
165
Unlike earlier keys, these two keys distinguish physiological stages of ovarian
development, not just external secondary sex characteristics. The difference allows
for more detailed analysis of the reproductive cycle. For example, the key of
Makarov & Denys (1980), commonly used for E. superba, does not distinguish
between females in previtellogenesis and those undergoing ovarian reorganisation,
i.e. females at the beginning and end of their reproductive cycle. Nor do they
separate females in late vitellogenesis from females in the final stages of egg
maturation. Finally the Makarov & Denys key does not distinguish females continuing to produce eggs after a spawn from those at the end of their reproductive
season.
Analysis of sexual development stages in the populations of E. superba west of the
Antarctic Peninsula and M. norvegica in the Ligurian Sea in the Mediterranean has
led to an understanding of the timing and duration of the different stages of ovarian
maturation (Fig. 6.5). The pattern of ovarian development in the two species is
basically similar. In both species previtellogenesis occurs in the whole ovary prior to
the start of the reproductive season, January to late February in M. norvegica, and
for a longer period from September through November in E. superba. Vitellogenesis in both species is cyclical, with successive batches of ooÈcytes brought to
maturity and released throughout the reproductive cycle in the spring for M. norvegica or summer for E. superba. The energy source appears to be different during
the two phases of yolk accumulation. In E. superba, previtellogenesis cannot continue without an outside source of energy because this species does not store enough
lipid for this process. In both species, the fat body in previtellogenic krill becomes
progressively better developed (Cuzin-Roudy, 1993). Since vitellogenesis and egg
maturation continue even with poor food, the `fat body' may be a source of lipid
during this time and reduce dependence on immediately available food during
vitellogenesis. The limited period of egg production alternates with a longer period
of gonadal rest.
However, there are some differences between the two species. In M. norvegica,
the ovary continually produces new yoc throughout the season, and ooÈcyte development is progressive; this strategy leads to rapid successive cycles of small batches
of embryos. In E. superba, production of yoc is not constant. The yoc at the start of
ovarian maturation mature in successive batches, with the possibility of four different stages of ooÈcytes (mature, vitellogenic, previtellogenic, and yoc) simultaneously in the ovary. The strategy is one of pulses of ooÈcyte production in spawning
episodes with large batches of eggs at the individual level. At the population level,
both strategies lead to more even egg production over a 2 to 4 month period. This
dispersal of eggs and larvae in time and space will enhance survival of the young in
variable environments. A similar cycle was inferred for T. inermis in the Bay of
Fundy from descriptions of ooÈcyte types and combinations in the ovary throughout
the season (Kulka & Corey, 1978). We suspect that this general pattern is common
to all euphausiids, but additional studies of the sexual development stages of other
species will be necessary.
Krill: Biology, Ecology and Fisheries
52
Cy
cl e
166
ve
51
May
4
June
1
Reproductive Rest
Re
pro
du
cti
6
Meganyctiphanes norvegica
Ligurian Sea
1 Feb
3
Dec
2
Onset
Spawning
Rest
Re
Cycle
e
tiv
uc
od cle
Cy
pr
5
4
6
7
Dec
3
Sept
8
Apr
2
Re
pr
od
9
1
uc
tiv
e
10
Re
st
Euphausia superba
west Ant Penin
Fig. 6.5 Alternation of reproductive cycle (black circles) and resting phases (pale grey stars)
throughout the year for (a) the Northern krill, Meganyctiphanes norvegica, in the Ligurian Sea of
the Mediterranean, redrawn from Cuzin-Roudy & Buchholz (1999); and (b) the Antarctic krill,
Euphausia superba, from samples west of the Antarctic Peninsula (data of Quetin & Ross).
Numbers in circles refer to sexual development stages as described in Tables 6.4 and 6.5. Onset
means onset of ovarian maturation or previtellogenesis. Spawning is period when eggs are released.
The arrows indicate stages involved in ovarian cycling.
Reproduction in Euphausiacea
167
Population fecundity
Population fecundity, i.e. total rate of egg production of the population in a
region, and early larval survivorship are the two main factors affecting recruitment in a region. Population fecundity in euphausiids is a product of female
abundance, the percentage of females reproducing that season, the number of
spawning episodes, and the number of eggs per spawning episode. The number of
spawning episodes is a function of the duration of the spawning season and how
often the ovary recycles during that period, i.e. the length of time between
spawning episodes.
Percentage population reproducing
The assumption has been that all females, once they are the size of a mature individual, will reproduce during that reproductive season, i.e. Siegel & Loeb (1994) for
E. superba. An examination of the sexual development stages present in the
population is a test of that assumption. If all females larger than the size at maturity
(Table 6.1) are in late previtellogenesis, vitellogenesis or maturation during the
reproductive season, then the entire population is reproducing. However, if a percentage of the population is in a resting stage, then not all females will reproduce
that season. The assumption has been verified for M. norvegica in the Ligurian Sea
(Cuzin-Roudy & Buchholz, 1999), but refuted for E. superba west of the Antarctic
Peninsula (Shaw, 1997). In the case of E. superba, the percentage of the population
reproducing varies from less than 20% to nearly 100%. The percentage is the
proportion of females in the reproductive cycle or having just completed the
reproductive cycle of the total population of subadult and adult females in January.
Thus for E. superba the number of seasons that an individual reproduces is an
important variable. For a species that is long-lived, and which can reproduce for
several seasons, the strategy of delayed reproduction or putting energy into growth
instead of reproduction for a season can potentially enhance the total number of
offspring produced. However, most euphausiid species only reproduce during one
year (Table 6.1), with a small percentage surviving to reproduce over a second
season, so this phenomenon is probably rare.
Individual fecundity
Based on the results of spawning frequency experiments with E. pacifica in Puget
Sound, Ross et al. (1982) first proposed multiple spawning episodes in an individual
euphausiid. Subsequently, studies of ovarian cell structure and development, and
experiments with live female euphausiids of many species have established that
multiple spawning episodes or batches of eggs are common in euphausiids (Table
6.6). Multiple spawning episodes in a single season theoretically increases the
probability of reproductive success in a fluctuating environment.
168
Krill: Biology, Ecology and Fisheries
Table 6.6 Euphausiid species with multiple broods in a reproductive season, confirmed or
inferred from experiments or histological examination of the ovary or for N. couchii inferred from
number of cohorts per year.
Species
Expt
Histology
Reference
E. crystallorophias
X
Harrington & Thomas (1987)
E. hanseni
X
X
Stuart & Nicol (1986); Stuart (1992)
E. lucens
X
X
Stuart & Nicol (1986)
E. pacifica
X
E. superba
X
X
Ross & Quetin (1983); Cuzin-Roudy (1987)
M. norvegica
X
X
Cuzin-Roudy (1993); Cuzin-Roudy &
Buchholz (1999)
N. australis
X
X
Hosie & Ritz (1983)
N. capensis
X
Stuart & Nicol (1986)
N. couchii
infer
T. inermis
X
Ross et al. (1982)
Lindley (1982)
Kulka & Corey (1978)
The number of spawning episodes per female is a function of the duration of the
spawning season and the interbrood period, itself a function of the rate of ooÈcyte
development. For those species for which we have estimates, the interbrood period
ranges from a few days to 30 days (Table 6.7). Because of its dependence on ooÈcyte
development rate, the interbrood period (number of spawning episodes) is likely to
be a function of food availability, and thus will vary inter-annually and geographically. For E. superba, the interbrood period varies with location and year
(Quetin et al., 1994), with a minimum of 6 days and a maximum of 50 days in midsummer. With data on the duration of the spawning season, and estimates of
interbrood periods from spawning frequency experiments, estimates of the number
of spawning episodes per season for a reproducing E. superba range from 3 to 9 per
season (Table 6.7). Observations of individual spawning females held in the
laboratory for more than several days must be qualified with the possibility that
spawning frequency is impacted by the nutritional conditions in the laboratory.
However, these observations do suggest that individuals and populations may not
show the same patterns. For E. superba (Harrington & Ikeda, 1986; Nicol, 1989), not
all females produced second or third batches of embryos. Also the interval between
broods was often shorter than predicted from the spawning frequency of a group of
females (6±12 days, Harrington & Ikeda, 1986). In M. norvegica (Cuzin-Roudy &
Buchholz, 1999), spawning events in individuals were only separated by 1±2 days,
but spawning was confined to a restricted part of the moult cycle, which would
increase estimates of population interbrood periods. In a detailed study of the
relationship between spawning and moulting cycles, Cuzin-Roudy & Buchholz
(1999) showed that M. norvegica released a complete batch of eggs (one vitellogenic
673, 650
728
580, 460
600 (555±630)
556±803
E. crystalloophias
E. lucens
E. pacifica
E. superba
378±405
N. australis*
8±10 per life
2 to 4 vit cycle
if repro, 3±9
I: 2±6
L: 11±30
3±10 per life vs 135
per life
No. SPE per
season
30 d
2±3/IMP
5±33 d
2±10 d
±
2d
6±10 d
InterBat
period
650 per vit cycle
1309
1417
2809
1372
45 mm±2469; 50 mm±3588
45 mm±2900 OvEggs '82
45 mm±1700 OvEggs '78
< 16 mm, 60;
> 19 mm 132
17±138
14±28 d±1 fem±1
187
Av no. egg per fem
per SPE
49±193
3 loc: 513, 681, 735
1140±1688
627±3115
89±6167
263±3662
500±8000
OvEggs = 715770 + 415.1 (TL) '82
OvEggs = 77224 + 198.5 (TL) '78
12±296
No = 39.07 + 7.87 DWT
No = 0.00064TL^4.25
2±80
Range no. egg per SPE
Sources of information listed by species initials. E. c.: Harrington & Thomas (1987), Ideda (1986). E. l.: Pillar & Stuart (1988) lower estimate of SPE per lifespan, Stuart &
Nicol (1986), Stuart (1992) higher estimate of SPE per lifespan. E. p.: Brinton (1976), Ross et al. (1982), Suh et al. (1993), Iguchi & Ikeda (1994). E. s.: Cuzin-Roudy (1987),
Kikuno (1981), Kikuno & Kawamura (1983), Siegel (1985), Harrington & Ikeda (1986), Ross & Quetin (1983), Nicol (1989). M. n.: Marschall (1983), Cuzin-Roudy &
Buchholz (1999). N. a.: Hosie & Ritz (1982, 1983).
* egg size in ovisac
722, 740
M. norvegica
3 sites
[560±590]
Size egg mm
Species
Table 6.7 Size of egg and individual fecundity estimates for euphausiid species: number of spawning episodes (SPE) per season, interbatch period,
average number of eggs per female per SPE or ovarian eggs (OvEggs, E.s.) or vitellogenic cycle (M.n.), range in number of eggs per SPE or equation
relating number of eggs (either per SPE or OvEggs) and total length (TL). I = intermittent range, L = limited range for E. p.
Reproduction in Euphausiacea
169
170
Krill: Biology, Ecology and Fisheries
cycle) in two main and usually one residual spawning event during the premoult
period. In the Ligurian Sea, spawning moult cycles alternate with vitellogenic moult
cycles. This pattern may not hold throughout its distribution range; in areas of low
food availability, an extended period of reorganisation of the ovary may be
necessary before a new cycle of egg production begins. Furthering our understanding of what drives this variability is critical to understanding reproductive
success in euphausiids.
The number of eggs released per spawning episode varies significantly both
within and between species (Table 6.7). Although an increase in fecundity with
female size is predicted, the correlation with total length is usually quite low (i.e.
E. superba as cited in Ross & Quetin, 1983; Harrington & Ikeda, 1986; Nicol et
al., 1995). Most investigators have inferred that the number of eggs released per
spawning episode is influenced by factors other than body size. Somers (1991)
examined five different regression models for the relationship between fecundity
and female size for crustaceans. He concluded that the allometric model (logtransformed fecundity versus log-transformed female length) was the better
choice, primarily because of the relative scale independence and the ability to
make comparisons among taxa. From general allometric concepts, the relation
between fecundity and female volume (expressed as length cubed) should be
linear (i.e. two volume measures), as has been shown for the relationship between
wet weight and ovarian ooÈcytes in E. superba and M. norvegica (Xuefeng &
Rong, 1995; Cuzin-Roudy, 2000). The relationship between fecundity and size
may also show high variability because the numbers of eggs released by the same
female in successive spawning episodes are not equal, as seen in both M. norvegica and E. superba (Harrington & Ikeda, 1986; Nicol, 1989; Cuzin-Roudy &
Buchholz, 1999). In fact, if all mature or maturing ooÈcytes (oc4) in the ovary are
counted, the relationship between the number of eggs and the size of the
euphausiid improves significantly for both E. superba (Siegel, 1985; Xeufeng &
Rong, 1995; Cuzin-Roudy, 2000) and M. norvegica (Cuzin-Roudy, 2000). The
implication is that a batch of eggs is released over a period of days in several
spawning episodes, and that the sum of the eggs released in this suite of successive episodes correlates well with body size. Batch size or number of eggs per
spawning episode can also vary among geographical sites (Harrington & Ikeda,
1986) or between years (Siegel, 1985; unpublished data, Ross & Quetin). Presumably the nutritional conditions prior to and during the spawning season also
affect the size of the batches.
Mauchline (1988) found that the relationship between brood volume and body
volume for 13 species of brooding euphausiids was logarithmic, with a correlation of
0.871. The slope was slightly greater than 1, implying that larger species tended to
produce larger broods, but the increase was small, from 10 to 15% of body volume.
For six species of epipelagic euphausiids, the relationship between the average
number of eggs per spawning episode or vitellogenic cycle and the volume (maximum total length cubed) was linear, with r 2 = 0.987 (Fig. 6.6a).
Reproduction in Euphausiacea
171
Fig. 6.6 Intraspecific comparison of euphausiid volume (expressed as the cube of total maximum
length, Table 6.1) and (a) average number of eggs per spawning episode, Table 6.7; and (b) average
diameter of the egg, average of values shown in Table 6.7 plus E. n. from Hirota et al. (1984). Initials
on points are the first letter of the genus and species of euphausiid. The line shown in (a) is average
number of eggs = 32.4 + 7.58 . TL3; n = 6, r2 = 0.987.
Embryo size
Embryo size also varies within and between species. Inter-annual and geographical
differences in embryo size are well documented for euphausiids (Mauchline &
Fisher, 1969). Such differences have implications for the energy reserves available
for the non-feeding larvae, and may impact survivorship of early life history stages.
172
Krill: Biology, Ecology and Fisheries
No studies have explicitly addressed this question, but variation in parental input
into the ooÈcyte before release and its impact on inter-annual variability in recruitment should be investigated.
Mauchline (1988) compared average egg volume and body volume for a suite of
16 epipelagic euphausiids, 7 brooders and 9 broadcast spawners. The relationship
was allometric, with r = 0.833:
log egg volume (ml) = 0.301 log body volume (ml) = 4.165
The exponent less than unity suggests that large euphausiids have larger eggs than
small euphausiids, but that egg size relative to body size decreases. From a slightly
different perspective, however, the conclusions are somewhat different. A plot of
egg volume against body volume (maximum TL cubed) shows that egg volume does
increase with body volume for six of the eight species of euphausiids plotted.
However, the egg of E. lucens is larger than would be predicted, and that of E.
superba much smaller than would be predicted. Some interesting differences were
obscured by the use of a log-log plot.
Mauchline (1988) suggested that there was a minimum viable size of egg to
explain the decrease in relative egg size with increasing body size. The hypothetical
minimum egg size also has implications for individual fecundity. Euphausiids can
increase their total fecundity either by producing successive broods, as has been
shown to be true for most euphausiids, or by decreasing egg size relative to body size
in order to increase the number of eggs per spawning episode. In the case of E.
superba, it appears as if the overall reproductive strategy includes both mechanisms.
6.3
Development
Eggs from different species vary not only in absolute size but also in the size of the
perivitelline space (PVS), the space between the vitelline membrane and the capsule. The duration and extent of the ontogenetic migration during the early life
history of euphausiids is influenced by the size of the PVS, and sinking rate and
development rate of the embryo exposed to characteristic hydrographic conditions
(Ross & Quetin, 1991; Hofmann et al., 1992). Many investigators have suggested
that a strong developmental descent and ascent is characteristic of oceanic
euphausiids (Makarov, 1979, 1983; Mauchline, 1980; Marr, 1962; Hempel et al.,
1979; Williams & Lindley, 1982). These species would have small PVS. Neritic
species, on the other hand, would have a large PVS, and would be either neutrally
buoyant or have slow sinking rates. The vertical extent of the developmental descent and ascent would be small. The larger vertical extent in oceanic euphausiids
would reduce intraspecific predation, whereas the small vertical extent in neritic
euphausiids would reduce the extent of benthic predation. The data to date (Table
6.8) do not clearly support this hypothesis. For M. norvegica and T. raschii, both
Reproduction in Euphausiacea
173
Table 6.8 Perivitelline space (PVS) as percentage volume of embryo and qualitative characterisation, sinking rate and reference.
Species
Neritic
E. crystallorophias
PVS %
vol
±
Qual char of
PSV
Sink rate
(m d±1)
Reference
`larger than
E. superba
`neutral'
Harrington & Thomas
(1987)
Marschall (1983)
Mauchline & Fisher (1969)
Marschall (1983)
Mauchline & Fisher (1969)
M. norvegica
78
T. raschii
52
Oceanic
E. pacifica
E. superba
±
9
small
small
E. nana
T. longicaudata
±
±
small
`very small'
`med-large'
`large'
61±104
132±180
42±67
132±180
Sink
203±231
175
±
±
Bollens et al. (1992)
Marschall (1983)
Quetin & Ross (1984)
Williams & Lindley (1982)
slow and fast sinking rates have been measured (Mauchline & Fisher, 1969; Marschall, 1983). For some species with a small PVS, high sinking rates but fast
development rates still yield a small vertical excursion, as seen in E. pacifica in
Dabob Bay (Bollens et al., 1992). More research is clearly warranted.
6.4
Implications for fisheries management
The goal of management of euphausiid fisheries, as for all fisheries, is to ensure that
recruitment to the adult population is adequate to compensate for the stock that is
harvested. The harvest rate must depend on the rate of recruitment into the adult
population that in turn depends on the population fecundity. Thus reproduction
(egg production) and its impact on recruitment should be a critical aspect of management of crustacean fisheries (Botsford, 1991). However, although recruitment
depends on egg production through the relationship between stock and recruitment,
the impact of variation in egg production is subsequently modified by the impact of
the fluctuating pre-recruitment environment on the survival of the early life history
stages.
At present we know very little about stock-recruitment relationships in crustaceans (Botsford, 1991). Data necessary to establish the relationship between
spawning stock and reproductive output in euphausiids are scarce. The time series
needs to be long, and both stock and recruitment must be measured on relevant time
and space scales. One critical issue is whether the stock measured is the one that
produced the larvae recruited to that adult population, or if the eggs and larvae were
transported by currents prior to recruitment to the adult population.
Zelikman et al. (1980) suggested that the lack of correlation between stock and
174
Krill: Biology, Ecology and Fisheries
recruitment in euphausiids supported the concept of a `self regulating' population,
i.e. density dependent mechanisms regulating recruitment. The amount of uncertainty about the relative importance of egg production and early larval mortality on
recruitment, however, suggests that this conclusion may be premature. The effects
of egg production on recruitment may be hidden by processes occurring during the
early life history stages, i.e. the effects of abiotic and biotic factors in the prerecruitment environment. In the case of E. superba, we know that interannual
variability in recruitment success of E. superba is high (Siegel & Loeb, 1995).
Although the mechanism proposed is the over-winter survival of the larvae (Ross &
Quetin, 1991; Quetin et al., 1996; Siegel & Loeb, 1995), the role of reproductive
output has not been thoroughly evaluated.
An alternative explanation for the lack of correlation between stock and
recruitment is that population fecundity is complex and varies greatly with stock.
Emerging evidence for euphausiids suggests that the number of eggs produced per
female may show large inter-annual variation due to the effect of nutritional history
on factors such as the percentage of the population reproducing, variation in
numbers of spawning episodes and batch size. Early larval survival may also be
affected by the amount of reserves in the embryo. Variation in embryo size for
euphausiids is well documented, and this variation implies that the amount of
reserves also varies.
Given our uncertainties about the relative roles of egg production and larval
survivorship in recruitment, the conservative management tactic would be to guard
egg production. Botsford (1991) presents the argument that such a management
tactic does not always create the highest yield, but that the population will be
maintained. To achieve this goal of guarding egg production, restrictions may be
placed on the timing or spatial occurrence of commercial harvesting, or on the part
of the population that can be fished. One common tactic when reproduction is
concentrated in one part of the year is a limited season for harvesting after the
reproductive season. With the species of euphausiids currently harvested, the
limited reproductive seasons would allow this tactic.
The restricted distribution of spawning populations would allow the use of a
somewhat different policy, basically to prohibit the harvesting of gravid females,
analogous to prohibiting `berried' lobsters and crabs from being kept. Fishing
pseudopopulations, for example, would have no immediate impact on population
recruitment. We also know that swarming is often correlated with reproductive
activity, so fishing surface swarms may selectively harvest the portion of the
population that we most want to guard. On the other hand, if we understand the
behaviour of individuals in the surface aggregations better, we may be able to take
advantage of differences in the behaviour of mated females in terms of timing of
spawning or diel distribution.
A more complex management tactic would be based on the fact that population
fecundity will depend on the size±age structure of the population, which in turn
depends on harvest policies. Larger females produce the greater number of eggs.
Reproduction in Euphausiacea
175
The goal would be to prevent the overfishing of the large egg-producing females and
the males needed to mate with those females throughout the season. In some
crustacean fisheries, a size limit is set for a single sex fishery (e.g. males below a
certain size, or females above a certain size), but this tactic would not be possible
with most euphausiids. Even though catches often show a skewed sex ratio (Watkins
et al., 1992) throwing back the `wrong' sex would not be feasible. Once caught, large
individuals would not survive if thrown back. In summary the best options for
management tactics based on our knowledge of euphausiid reproduction appear to
be restrictions of the fishery in time and space to guard the large egg-producing
females.
Acknowledgments
We thank our many colleagues in the field of euphausiid research for stimulating
and enlightening discussions. Those with J. Cuzin-Roudy and S. Nicol have been
particularly valuable. An anonymous reviewer asked questions that helped clarify
the text. This publication was supported by NSF Grant OPP96-32763, with additional support from the University of California at Santa Barbara. This is Palmer
LTER publication No. 182.
References
Astthorsson, O.S. (1990) Ecology of the Euphausiids Thysanoessa raschi, T. inermis and
Meganyctiphanes norvegica in Isafjord-deep, northwest-Iceland. Mar. Biol. 107, 147±57.
Baker, A.C. (1959) The distribution of Euphausia triacantha Holt and Tattersall. Discovery
Rep. 29, 309±339.
Bargmann, H.E. (1937) The reproductive system of Euphausia superba. Discovery Rep. 14,
325±50.
Bargmann, H.E. (1945) The development and life-history of adolescent and adult krill,
Euphausia superba. Discovery Rep. 23, 103±176.
Blackburn, M. (1980) Observations on the distribution of Nyctiphanes australis Sars (Crustacea, Euphausiidae) in Australian waters. Commonwealth Scientific and Industrial
Research Organization Division of Fisheries and Oceanography, Cronulla, N.S.W.
Bollens, S.M., Frost, B.W. & Lin, T.S. (1992) Recruitment, growth, and diel vertical migration
of Euphausia pacifica in a temperate fjord. Mar. Biol. 114, 219±28.
Botsford, L.W. (1991) Crustacean egg production and fisheries management. In Crustacean
Egg Production, Vol. 7 (A. Wenner & A. Kuris, eds), pp. 379±94. Balkema, Rotterdam.
Boysen, E. & Buchholz, F. (1984) Meganyctiphanes norvegica in the Kattegat. Studies on the
annual development of a pelagic population. Mar. Biol. 79, 195±207.
Brinton, E. (1976) Population biology of Euphausia pacifica off southern California. Fish.
Bull. 74, 733±62.
Brinton, E. & Townsend, A. W. (1991) Developmental rates and habitat shifts in the Antarctic neritic euphausiid Euphausia crystallorophias, 1986±87. Deep-Sea Res. 38, 1195±211.
176
Krill: Biology, Ecology and Fisheries
Cuzin-Roudy, J. (1987) Sexual differentiation in the Antarctic krill Euphausia superba Dana
(Crustacea: Euphausiacea). J. Crust. Biol. 7, 518±24.
Cuzin-Roudy, J. (1993) Reproductive strategies of the Mediterranean krill, Meganyctiphanes
norvegica and the Antarctic krill, Euphausia superba (Crustacea, Euphausiacea). Invertebrate Repro Devel. 23, 105±114.
Cuzin-Roudy, J. (2000) Seasonal reproduction, multiple spawning and fecundity in Northern
krill, Meganyctiphanes norvegica, and Antarctic krill, Euphausia superba. In Proceedings of
the Second International Krill Symposium, Santa Cruz, California, August 1999. Can. J.
Fish. Aquat. Sci. (in press).
Cuzin-Roudy, J. & Amsler, M.O. (1991) Ovarian development and sexual maturity staging in
antarctic krill, Euphausia superba Dana (Euphausiacea). J. Crust Biol. 11, 236±49.
Cuzin-Roudy, J. & Buchholz, F. (1999) Ovarian development and spawning in relation to the
moult cycle in Northern krill, Meganyctiphanes norvegica (Crustacea: Euphausiacea),
along a climatic gradient. Mar. Biol. 133, 267±81.
Cuzin-Roudy, J. & Labat, J.P. (1992) Early summer distribution of Antarctic krill sexual
development in the Scotia-Weddell region: a multivariate approach. Polar Biol. 12, 65±
74.
Dalpadado, P. & Skjoldal, H.R. (1996) Abundance, maturity and growth of the krill species
Thysanoessa inermis and T. longicaudata in the Barents Sea. Mar. Ecol. Prog. Ser. 144, 175±
83.
Einarsson, H. (1945) Euphausiacea. I. Northern Atlantic species. Dana Report 27, 1±175.
Endo, Y. (1984) Daytime surface swarming of Euphausia pacifica (Crustacea: Euphausiacea)
in the Sanriku coastal waters off northeastern Japan. Mar. Biol. 79, 269±76.
Falk-Petersen, S. & Hopkins, C. C. E. (1981) Ecological investigations on the zooplankton
community of Balsfjorden, northern Norway: population dynamics of the euphausiids
Thysanoessa inermis (KroÈyer), T. raschii (M. Sars) and Meganyctiphanes norvegica (M.
Sars) in 1976 and 1977. J. Plank. Res. 3, 177±92.
Falk-Petersen, S., Hagen, W., Kattner, G., Clarke, A. & Sargent, J. (1999) Lipids ± Key
components of dominant species of Arctic and Antarctic krill. Paper presented at the Second
International Symposium on Krill, Santa Cruz CA, August 23±27, 1999.
Fevolden, S.E. (1980) Krill off Bouvetoya and in the southern Weddell Sea with a description
of larval stages of Euphausia crystallorophias. Sarsia 65, 149±62.
Foxton, P. (1956) The distribution of the standing crop of zooplankton in the Southern
Ocean. Discovery Rep. 28, 191±236.
Hanamura, Y., Kotori, M. & Hamaoka, S. (1989) Daytime surface swarms of the Euphausiid
ThysanoÈessa inermis off the west. Mar. Biol. 102, 369±76.
Harrington, S.A. & Ikeda, T. (1986) Laboratory observations on spawning, brood size and egg
hatchability of the antarctic krill Euphausia superba from Prydz Bay, Antarctica. Mar. Biol.
92, 231±35.
Harrington, S.A. & Thomas, P.G. (1987) Observations on spawning by Euphausia crystallrophias from waters adjacent to Enderby Land (East Antarctica) and speculations on the
early ontogenetic ecology of neritic euphausiids. Polar Biol. 7, 93±95.
Hart, T.J. (1942) Phytoplankton periodicity in Antarctic surface waters. Discovery Rep. 8, 1±
268.
Heath, W.A. (1977) The ecology and harvesting of euphausiids in the Strait of Georgia.
University of British Columbia, Vancouver, pp. 187.
Reproduction in Euphausiacea
177
Hempel, I., Hempel, G. & Baker, A.D. (1979) Early life history stages of krill (Euphausia
superba) in Bransfield Strait and Weddell Sea. Meereforschung 27, 267±81.
Hirota, Y., Nemoto, T. & Marumo, R. (1984) Larval development of Euphausia nana
(Crustacea: Euphausiacea). Mar. Biol. 81, 311±22.
Hofmann, E.E., Capella, J.E., Ross, R.M. & Quetin, L.B. (1992) Models of the early life
history of Euphausia superba ± Part I. Time and temperature dependence during the
descent±ascent cycle. Deep-Sea Res. 39, 1177±1200.
Hollingshead, K.W. & Corey, S. (1974) Aspects of the life history of Meganyctiphanes norvegica (M. Sars), Crustacea (Euphausiacea), in the Passamaquoddy Bay. Can. J. Zool. 52,
495±505.
Hosie, G.W. & Cochran, T.G. (1994) Mesoscale distribution patterns of macrozooplankton
communities in Prydz Bay, Antarctica ± January to February 1991. Mar. Ecol. Prog. Ser.
106, 21±39.
Hosie, G. & Ritz, D. (1983) Contribution of moulting and eggs to secondary production in
Nyctiphanes australis (Crustacea: Euphausiacea). Mar. Biol. 77, 215±20.
Iguchi, N. & Ikeda, T. (1994) Experimental study on brood size, egg hatchability and early
development of a euphausiid Euphausia pacifica from Toyama Bay, southern Japan Sea.
Bull. Japan Sea Nat. Fish. Res. Inst. 44, 49±57.
Iguchi, N., Ikeda, T. & Imamura, A. (1993) Growth and life cycle of a euphausiid crustacean
(Euphausia pacifica) Hansen in Toyama Bay, southern Japan Sea. Bull. Japan Sea Nat.
Fish. Res. Inst. 43, 69±81.
Ikeda, T. (1986) Preliminary observations on the development of the larvae of Euphausia
crystallorophias Holt and Tattersall in the laboratory. Mem. Natl Inst. Polar Res. Tokyo,
pp. 183±86.
Kikuno, T. (1981) Spawning behaviour and early development of the antarctic krill,
Euphausia superba Dana, observed on board R.V. Kaiyo Maru in 1979±80. Antarctic Rec.
73, 97±102.
Kikuno, T. & Kawamura, A. (1983) Observation of the ovarian eggs and spawning habits in
Euphausia superba Dana. Mem. Natl Inst. Polar Res. Tokyo. Special Issue 27, 104±128.
Kotori, M. (1995) An incidence of surface swarming of Euphausia pacifica off the coast of
western Hokkaido, Japan. Bull. Plank. Soc. Japan 42, 80±84.
Kulka, D.W. & Corey, S. (1978) The life history of Thysanoessa inermis (Kroyer) in the Bay
of Fundy. Can. J. Zool. 56, 492±506.
Lascara, C.M., Hofmann, E.E., Ross, R.M. & Quetin, L.B. (1999) Seasonal variability in the
distribution of Antarctic krill, Euphausia superba, west of the Antarctic Peninsula. DeepSea Res. 46, 951±84.
Lindley, J.A. (1978) Population dynamics and production of euphausiids. I. Thysanoessa
longicaudata in the North Atlantic Ocean. Mar. Biol. 46, 121±30.
Lindley, J.A. (1982) Population dynamics and production of Euphausiids. III. Meganyctiphanes norvegica and Nyctiphanes couchii in the North Atlantic Ocean and the North
Sea. Mar. Biol. 66, 37±46.
Makarov, R.R. (1977) Distribution of the larvae and some questions of the ecology of
reproduction of the euphausiid Euphausia frigida Hansen, 1911 (Crustacea, Euphausiacea)
in the southern part of the Scotia Sea. Oceanology 17, 208±213.
Makarov, R.R. (1979) Larval distribution and reproductive ecology of Thysanoessa macrura
(Crustacea: Euphausiacea) in the Scotia Sea. Mar. Biol. 52, 377±86.
178
Krill: Biology, Ecology and Fisheries
Makarov, R.R. (1983) Some problems in the investigation of larval euphausiids in the Antarctic. Berichte zur Polarforschung 4, 58±69.
Makarov, R.R. & Denys, C.J. (1980) Stages of sexual maturity of Euphausia superba Dana.
BIOMASS Handbook 11, 1±11.
Makarov, R.R. & Men'shenina, L.L. (1992) Larvae of euphausiids off Queen Maud Land.
Polar Biol. 11, 515±23.
Makarov, R.R. & Sysoyeva, M.V. (1985) Biology and distribution of Euphausia superba in
the Lazarev Sea and adjacent waters. Legkaya i pishevaya promyshlennost', Moscow, 110±
116.
Makarov, R., Men'shenina, L. & Spiridonov, V. (1990) Distributional ecology of euphausiid
larvae in the Antarctic Peninsula region and adjacent waters. Proceedings of the NIPR
Symposium on Polar Biology 3, 23±35.
Makarov, R.R., Men'shenina, L.L., Timonin, V.P. & Shurunov, N.A. (1991) Ecology of larvae
and reproduction of Euphausiidae in the Ross Sea. Soviet J. Mar. Biol. 16, 156±62.
Makarov, R.R., Solyankin, Y. & Shevtsov, V.V. (1985) Environmental conditions and
adaptive features of the biology of Euphausia superba Dana in the Lazarev Sea. Polar
Geogr. Geol. 9, 146±64.
Marr, J.W.S. (1962) The natural history and geography of the Antarctic krill (Euphausia
superba Dana). Discovery Rep. 32, 33±464.
Marschall, H.-P. (1983) Sinking speed, density and size of Euphausiid eggs. Meeresforschung
30, 1±9.
Mauchline, J. (1968) The development of the eggs in the ovaries of euphausiids and estimation of fecundity. Crustaceana 14, 155±63.
Mauchline, J. (1980) The biology of mysids and euphausiids. Adv. Mar. Biol. 18, 1±681.
Mauchline, J. (1988) Egg and brood sizes of oceanic pelagic crustaceans. Mar. Ecol. Prog. Ser.
43, 251±58.
Mauchline, J. & Fisher, L.R. (1969) The biology of euphausiids. Adv. Mar. Biol. 7, 1±454.
Men'shenina, L.L. (1988) Ecology of spawning and larval development of Thysanoessa
macrura in the southwestern sector of the Antarctic Ocean (with remarks on T. vicina).
Oceanology 28, 774±80.
Men'shenina, L.L. (1992) Distribution of Euphausiid larvae in the Weddell Gyre in Sep.±Oct.
1989. Proceedings of the NIPR Symposium on Polar Biology 5, 44±54. Tokyo, National
Institute of Polar Research.
Naito, Y., Taniguchi, A. & Hamada, E. (1986) Some observations on swarms and mating
behavior of antarctic krill (Euphausia superba Dana). Mem. Natl Inst. Polar Res, Special
Issue 40, 178±82.
Nelson, K. (1991) Scheduling of reproduction in relation to molting and growth in malacostracan crustaceans. In Crustacean Egg Production, Vol. 7 (A. Wenner & A. Kuris, eds),
pp. 77±133. Balkema, Rotterdam.
Nemoto, T., Doi, T. & Nasu, K. (1981) Biological characteristics of krill caught in the
Southern Ocean. In Selected Contributions to the Woods Hole Conference on Living
Resources of the Southern Ocean 1976, Biomass Vol. 2., pp. 47±64. SCAR, Cambridge.
Nicol, S. (1984) Population structure of daytime surface swarms of the euphausiid
Meganyctiphanes norvegica in the Bay of Fundy. Mar. Ecol. Prog. Ser. 18, 29±39.
Nicol, S. (1989) Apparent independence of the spawning and moulting cycles in female
Antarctic krill (Euphausia superba Dana). Polar Biol. 9, 371±75.
Reproduction in Euphausiacea
179
Nicol, S. & Endo, Y. (1997) Krill fisheries of the world. FAO Fisheries Technical Paper 367,
100 pp. Food and Agriculture Organization of the United Nations, Rome.
Nicol, S., de la Mare, W.K. & Stolp, M. (1995) The energetic cost of egg production in
Antarctic krill (Euphausia superba Dana). Antarctic Sci. 7, 25±30.
Nicol, S., James, A. & Pitcher, G. (1987) A first record of daytime surface swarming by
Euphausia lucens in the Southern Benguela region. Mar. Biol. 94, 7±10.
O'Brien, D.P. (1988) Surface schooling behaviour of the coastal krill Nyctiphanes australis
(Crustacea: Euphausiacea) off Tasmania, Australia. Mar. Ecol. Prog. Ser. 42, 219±33.
Odate, K. (1991) Fishery biology of the krill, Euphausia pacifica, in the northeastern coasts of
Japan. Suisan Kenkyu Sosho 40, 1±100 (in Japanese).
Pakhomov, E.A. & Perissinotto, R. (1996) Antarctic neritic krill Euphausia crystallorophias:
spatio-temporal distribution, growth and grazing rates. Deep-Sea Res. 43, 59±87.
Paul, A.J., Coyle, K.O. & Ziemann, D.A. (1990) Timing of spawning of Thysanoessa raschii
(Euphausiacea) and occurrence of their feeding stage larvae in an Alaskan Bay. J. Crust.
Biol. 10, 69±78.
Pillar, S.C. & Stuart, V. (1988) Population structure, reproductive biology and maintenance of
Euphausia lucens in the southern Benguela Current. J. Plank. Res. 10, 1083±98.
Ponomareva, L.A. (1966) The euphausiids of the North Pacific, their distribution, ecology, and
mass species. Inst. Okeanol. Akad. Sci. USSR, Moscow.
Quetin, L.B. & Ross, R.M. (1984) Depth distribution of developing Euphausia superba
embryos, predicted from sinking rates. Mar. Biol. 79, 47±53.
Quetin, L.B., Ross, R.M. & Clarke, A. (1994) Krill energetics: seasonal and environmental
aspects of the physiology of Euphausia superba. In Southern Ocean Ecology: the BIOMASS
Perspective (S. El-Sayed, ed.), pp. 165±84. Cambridge University Press, Cambridge.
Quetin, L.B., Ross, R.M., Frazer, T.K. & Haberman, K.L. (1996) Factors affecting distribution and abundance of zooplankton, with an emphasis on Antarctic krill, Euphausia
superba. In Foundations for Ecological Research West of the Antarctic Peninsula, Vol. 70
(R.M. Ross, E.E. Hofmann, & L.B. Quetin, eds), pp. 357±71. American Geophysical
Union, Washington, D.C.
Quetin, L.B., Ross, R.M., PreÂzelin, B.B., Haberman, K.L., Hacecky, K.L. & Newberger, T.
(1992) Palmer LTER program: Biomass and community composition of euphausiids within
the peninsula grid, November 1991 cruise. Antarctic J. US 27, 244±45.
Ritz, D. & Hosie, G. (1982) Production of the euphausiid Nyctiphanes australis in Storm Bay,
south-eastern Tasmania. Mar. Biol. 68, 103±108.
Ritz, D.A. (1994) Social aggregation in pelagic invertebrates. In Advances in Marine Biology,
Vol. 30 (J.H.S. Blaxter & A.J. South, eds), pp. 155±207. Academic Press, London.
Ross, R.M. & Quetin, L.B. (1983) Spawning frequency and fecundity of the antarctic krill
Euphausia superba. Mar. Biol. 77, 201±205.
Ross, R.M. & Quetin, L.B. (1991) Ecological physiology of larval euphausiids, Euphausia
superba (Euphausiacea). Memoirs of the Queensland Museum 31, 321±33.
Ross, R.M., Daly, K.L. & English, T.S. (1982) Reproductive cycle and fecundity of Euphausia
pacifica in Puget Sound, Washington. Limn. Oceanogr. 27, 304±314.
Ross, R.M., Quetin, L.B., Amsler, M.O. & Elias, M.C. (1987) Larval and adult Antarctic krill,
Euphausia superba, winter-over at Palmer Station. Antarctic J. US 22, 205±206.
Shaw, C.T. (1997) Effect of sea ice conditions on physiological maturity of female Antarctic
krill (Euphausia superba Dana) west of the Antarctic Peninsula. Master of Arts thesis,
180
Krill: Biology, Ecology and Fisheries
Ecology, Evolution and Marine Biology, University of California at Santa Barbara, Santa
Barbara, pp. 98. Available through interlibrary loan, call # QH319.C2.S25 SHAC 1997, or
through the authors of this chapter.
Siegel, V. (1985) On the fecundity of Antarctic krill, Euphausia superba (Euphausiacea).
Archiv fuÈr FischereiWissenschaft 36, 185±93.
Siegel, V. (1986) Structure and composition of the Antarctic krill stock in the Bransfield Strait
(Antarctic Peninsula) during the Second International BIOMASS Experiment (SIBEX).
Archiv fuÈr FischereiWissenschaft 37, 51±72.
Siegel, V. (1987) Age and growth of antarctic Euphausiacea (Crustacea) under natural
conditions. Mar. Biol. 96, 483±95.
Siegel, V. (1992) Assessment of the krill Euphausia superba spawning stock off the Antarctic
Peninsula. Archiv fuÈr Fischereiwissenschaft 41, 101±130.
Siegel, V. & Loeb, V. (1994) Length and age at maturity of Antarctic krill. Antarctic Sci. 6,
479±82.
Siegel, V. & Loeb, V. (1995) Recruitment of Antarctic krill (Euphausia superba) and possible
causes for its variability. Mar. Ecol. Prog. Ser. 123, 45±56.
Siegel, V., Bergstrom, B., Stromberg, J.O. & Schalk, P.H. (1990) Distribution, size frequencies and maturity stages of krill, Euphausia superba, in relation to sea-ice in the
northern Weddell Sea. Polar Biol. 10, 549±57.
Smiles, M.C., Jr. & Pearcy, W.G. (1971) Size structure and growth rate of Euphausia pacifica
off the Oregon coast. Fish. Bull. 69, 79±86.
Smith, S.L. (1991) Growth, development and distribution of the euphausiids Thysanoessa
raschii (M. Sars) and Thysanoessa inermis (KroÈyer) in the southeastern Bering Sea. Polar
Res. 10, 461±78.
Smith, S.E. & Adams, P.B. (1988) Daytime surface swarms of Thysanoessa spinifera
(Euphausiacea) in the Gulf of the Farallones, California. Bulletin of Marine Science 42, (1),
76±84.
Somers, K.M. (1991) Characterizing size-specific fecundity in crustaceans. In Crustacean Egg
Production, Vol. 7 (A. Wenner & A. Kuris, eds), pp. 357±78. Balkema, Rotterdam.
Spiridonov, V.A. (1995) Spatial and temporal variability in reproductive timing of Antarctic
krill (Euphausia superba Dana). Polar Biol. 15, 161±74.
Stepnik, R. (1982) All-year populational studies of Euphausiacea (Crustacea) in the
Admiralty Bay (King George Island, South Shetland Island Antarctic). Polish Polar Res. 3,
49±68.
Stuart, V. (1992) Fecundity of Euphausia lucens (Hansen) ± Laboratory evidence for multiple
broods. J. Exp. Mar. Biol. Ecol. 160, 221±28.
Stuart, V. & Nicol, S. (1986) The reproductive potential of three euphausiid species from the
southern Benguela region. J. Exp. Mar. Biol. Ecol. 103, 267±74.
Stuart, V. & Pillar, S.C. (1988) Growth and production of Euphausia lucens in the southern
Benguela Current. J. Plank. Res. 10, 1099±112.
Suh, H.L., Soh, H.Y. & Hong, S.Y. (1993) Larval development of the euphausiid Euphausia
pacifica in the Yellow Sea. Mar. Biol. 115, 625±33.
Timofeyev, S.F. (1994) Population structure of summer aggregations of euphausiids Thysanoessa raschii (M. Sars) in the southern part of the Barents Sea. Oceanology 33, 785±89.
Virtue, P., Nichols, P.D., Nicol, S. & Hosie, G. (1996) Reproductive trade-off in male Antarctic krill, Euphausia superba. Mar. Biol. 126, 521±27.
Reproduction in Euphausiacea
181
Watkins, J.L., Buchholz, F., Priddle, J., Morris, D.J. & Ricketts, C. (1992) Variation in
reproductive status of Antarctic krill swarms ± Evidence for a size-related sorting
mechanism. Mar. Ecol. Prog. Ser. 82, 163±74.
Williams, R.L. & Lindley, J.A. (1982) Variability in abundance, vertical distribution and
ontogenetic migrations of Thysanoessa longicaudata (Crustacea: Euphausiacea) in the
North-Eastern Atlantic. Mar. Biol. 69, 321±30.
Xuefeng, Z. & Rong, W. (1995) Reproductive characteristics of Antarctic krill, Euphausia
superba Dana, in the Prydz Bay region. Antarctic Res. 6, 58±72.
Zelikman, E.A. (1958) On gonad maturation and female productivity in species of
euphausians abundant in the Barents Sea. Doklady Adad Nauk S.S.S.R 118, 201±204.
Zelikman, E.A., Lukashevich, I.P., Drobysheva, S.S. & Degtereva, A.A. (1980) Fluctuations
of the number of eggs spawned by Thysanoessa inermis Kr. and T. raschii (M. Sars)
(Euphausiacea) from the Barents Sea. Oceanology 20, 716±21.
Chapter 7
Role of Krill in Marine Food Webs
7.1
Japanese waters
Yoshi Endo
Off the Japanese coast Euphausia pacifica is preyed upon by almost all the commercially important fish species which are endemic to, or seasonally migrate
through Sanriku waters (e.g., Endo, 1981; Yamamura, 1993; Kuroda, 1994; Yamamura et al., 1998). It is one of the main food items of Pacific cod (Gadus macrocephalus), walleye pollack (Theragra chalcogramma), Japanese chub mackerel
(Scomber japonicus) and sand lance (Ammodytes personatus). In the E. pacifica
fishing season, sand lance, black-tailed gull (Larus crassirostris) and rhinoceros
auklet (Cerorhinca monocerata) feed on them intensively (Komaki, 1967). For
example, Takeuchi (1985) has reported that on average each individual sand lance
in a school contained 82 individual E. pacifica in its stomach. In this chapter, I
consider the importance of E. pacifica both in demersal and pelagic environments,
with the former being reported more fully.
Demersal environment
Fujita (1994) suggested that E. pacifica is most important as food for demersal fishes
in the upper slope region. There, compared with the continental shelf and deeper
areas (> 300 m) in Sanriku waters, the abundance of demersal fish is highest, but the
species diversity is low. Blue eye, Chlorophthalmus albatrossis (borealis), is a finfish
for which the annual commercial catch is small, about 200±400 tonnes. It feeds on E.
pacifica throughout the year in waters off the Fukushima Prefecture (Akiyama,
1996). This implies that E. pacifica occurs in the offshore area within the depth
range from 150 to 260 m throughout the year.
Yamamura (1993) examined the resource partitioning of demersal fish in the
offshore areas of Sendai Bay during May to June 1989 (see Fig. 3.1.4 for place
names). Pacific cod was dominant in areas where the water was shallower than 200
m deep and they made up 76% by number of the demersal fish present there.
Walleye pollack was dominant between 300±400 m accounting for 70% of the fish,
and threadfin hakeling (Laemonema longipes) was dominant, 89% numerically, of
182
Role of Krill in Marine Food Webs
183
all fish in the area deeper than 400 m. Stomach contents analysis showed that Pacific
cod with body length < 40 cm eat mainly E. pacifica and the amphipod (Themisto
japonica). Larger individuals eat Japanese sardine (Sardinops melanostictus).
Walleye pollack, irrespective of their body size, eat mainly E. pacifica. Threadfin
hakeling with body length < 40 cm eat copepods, euphausiids and mesopelagic
decapods. Larger individuals eat lantern fishes. It is noteworthy that 62% of total
biomass of demersal fishes were zooplankton feeders whereas benthos feeders
comprised only 1.3%. This implies that the relative abundance of pelagic prey is
high in this area.
Based on four years of bottom trawl surveys in a similar survey area off Sendai
Bay in May and November, Yamamura et al. (1998) examined the importance of E.
pacifica as food of demersal fishes. Of 87 fish species examined for stomach contents, 24 (27.6%) ingested E. pacifica. In terms of biomass, however, E. pacifica
feeders comprised more than 90% in 10 out of 16 sample sets. Pacific cod and
walleye pollack had the highest predation impact. The contribution of E. pacifica to
the total diet of the fish assemblages was highest in shallow (< 300 m) regions during
May, accounting for 40% by mass. Yamamura et al. (1998) found a smaller contribution of E. pacifica in warm years, and pointed out that the benthopelagic
population of E. pacifica moved offshore in accordance with the preferred temperature range. The average predation impact on E. pacifica was maximal in May
(4.6 kg wet wt km±2 d±1) and minimal in November in the shallow area (0.4 kg wet wt
km±2 d±l). Annual estimates of consumption by demersal fishes in the survey area
ranged from 15 to 64% of the annual commercial catch of E. pacifica by local
fisheries. The authors suggested that E. pacifica fisheries have the potential to exert
a considerable impact on demersal fish assemblages.
Kodama & Izumi (1994) compiled data on demersal fish landings at the Ishinomaki Fish Market, and examined stomach contents to quantify the importance of E.
pacifica as prey. These landings were made by an offshore trawl fishery which is
operated mainly in the vicinity of Kinkasan Island over water 150 to 2000 m deep.
The average yearly landing of demersal fishes at the Ishinomaki Fish Market was
45 740 tons during the period from 1983 to 1988. Except for 10 127 tons of sand lance
caught in the shallow Sendai Bay area, gadoid fishes predominated constituting
88% (31 355 tons) of the total landings. These were followed by Japanese common
squid (Todarodes pacificus) 1.8% (652 tons), spinycheek rockfish (Sebastolobus
macrochir) 1.3% (460 tons), octopuses 0.6% (210 tons) and flatfishes 0.6% (203
tons). Gadoid fish catches were dominated by walleye pollack which made up 76%
of the catch. This was followed by the other gadoids, Pacific cod 7%, and deep-sea
rattails 17%. This indicates the importance of walleye pollack as a demersal fish
within the ecosystem of this area.
The importance of E. pacifica as food is linked to the body size of the fish predators (Kodama & Izumi, 1994). For instance, the percentages of walleye pollack
which consumed E. pacifica as prey were 84% for fish of body length of 10±19 cm,
77% for fish of body length 20±29 cm, 60% for fish of body length 30±39 cm, and
184
Krill: Biology, Ecology and Fisheries
42% at 40±64 cm. For Pacific cod, comparable results were 24% at 10±19 cm, 47% at
20±29 cm, 13% at 30±39 cm, 3% at 40±59 cm and 0% at 60±79 cm. For Pacific
herring, E. pacifica were consumed by 65% of fish between 17±23 cm long and 70%
of fish 24±32 cm long.
Several assumptions must be made to calculate daily consumption of E. pacifica
by demersal fish. In developing their trophic analysis of the ecosystem, Kodama &
Izumi (1994) categorised the Pacific herring as a demersal fish because, over the
southern part of its range in Sanriku waters, it behaves like a demersal fish
(Kodama, personal communication). Three general assumptions were made
regarding the energy balance in the system. Firstly, fishing efficiency was assumed to
be 0.3 and therefore fish biomass equals annual catch divided by 0.3. Secondly, the
daily ration was assumed to be equal to the average weight of stomach contents.
This assumes that the stomach is filled no more than once a day, an assumption that
has to be made because the assimilation rate is unknown. This may cause an
underestimation of the amount of E. pacifica consumed. Thirdly, the ratio of E.
pacifica weight to the total weight of stomach contents was assumed to be equal to
the occurrence of E. pacifica divided by the sum of occurrences of all the food items.
The results are shown in Table 7.1.1. Walleye pollack are estimated to consume
540 000 tons, Pacific cod 8500 tons and Pacific herring 1300 tons of E. pacifica per
year. A total of 550 000 tons were consumed each year off Miyagi and Fukushima
Prefectures. As about 52% of demersal fish landings from the whole Sanriku coast
to Kashima-nada are landed at the Ishinomaki Fish Market, the consumption of E.
pacifica by fish probably amounts to about 1 000 000 tons per annum.
As discussed by Endo in Chapter 3, the benthopelagic population of E. pacifica
occurs in the daytime throughout the year except in spring in the Sanriku and Joban
Table 7.1.1 Estimates of Euphausia pacifica consumed annually by the three demersal fish species
in the vicinity of Kinkasan Island (Adapted from Kodama & Izumi, 1994).
Fish species
Body
length
(cm)
Annual
catch
(t)
Biomass
(t)
Walleye
pollack
10±25
25±35
> 40
total
12 265
18 215
10 346
40 826
Pacific cod
15±25
30±40
> 50
total
Pacific herring
1)
2)
1-E1)
Daily
ration
(%)
Krill
ratio2)
Annual Krill
consumption
(t)
40 883
60 717
34 487
136 087
0.904
0.935
0.891
2.3
1.7
1.5
0.74
0.69
0.44
231 594
237 339
75 503
544 436
496
1814
1067
3377
1 653
6047
3557
11 257
0.946
0.951
0.901
2.6
1.9
1.9
0.31
0.09
0.01
4542
3746
213
8501
180
600
0.641
1.5
0.61
1295
E denotes proportion of individuals with empty stomach.
Occurrence of Euphausia pacifica in the fish stomach relative to the occurrence of total food items.
Role of Krill in Marine Food Webs
185
areas. Benthopelagic E. pacifica support very high abundance of demersal fish
population. From submersible observations on benthopelagic E. pacifica we have
noted dense beds of brittle stars on the seabed (Endo & Kodama, in preparation). It
has been reported that the seas surrounding the northern part of Japan are
blanketed by dense beds of the brittle star Ophiura sarsii between depths of about
200 and 600 m with a mean density estimated to be 373 individuals m±2 (Fujita &
Ohta, 1989). The brittle stars may depend on E. pacifica and lantern fishes for food
(Stancyk et al., 1998). In addition to this information, quantitative estimates of the
biomass of benthopelagic E. pacifica are required for the entire distributional range
in Japanese waters to elucidate the relative importance of the species in the pelagic
as well as benthopelagic food web in Sanriku waters.
Pelagic environment
Pelagic fishes such as sardines, Japanese chub mackerel and Japanese common squid
also migrate in large numbers into the Sanriku area in May to January. Their biomass
is estimated to be more than 1 000 000 tons (Kodama & Izumi, 1994). Quantitative
estimates of E. pacifica consumed by these pelagic species are also needed.
Based on 40 years of coastal whaling operations from 1948 to 1987, Kasamatsu &
Tanaka (1992) reported that 46.5% of minke whale, Balaenoptera acutorostrata,
were feeding on krill in Joban±Sanriku waters. It is highly probable that most of the
krill were E. pacifica. Ogi (1994) has provided some information on seabirds which
feed on euphausiids in Sanriku waters. In early summer, the sooty shearwater,
(Puffinus griseus) the slender-billed shearwater (P. tenuirostris) and the pale-footed
shearwater (P. carneipes) visit Sanriku waters in large flocks during their northward
migration from the southern hemisphere.
Although the abundance of the rhinoceros auklet (Cerorhinca monocerata) and
the streaked shearwater (Calonectris leucomelas) which breed in this area, are
present in reasonable numbers, the sooty shearwater is by far the most numerous.
The slender-billed shearwater is more important as a euphausiid feeder than the
piscivorous sooty shearwater because the species is thought to be suited for sea
surface feeding (Ogi, 1994). This was confirmed by stable isotope analyses (Minami
et al., 1995). In Sanriku waters the slender-billed shearwater had significantly lower
d15N values than the sooty shearwater with a difference of 1.2%. This suggested that
the trophic level of the slender-billed shearwater was lower than that of the sooty
shearwater. In fact, as the slender-billed shearwater feed mainly on the euphausiid,
Nyctiphanes australis, in their breeding area of Tasmanian waters, Ogi (1994) suggests that they feed on a large amount of E. pacifica in Sanriku waters.
Euphausiids are reported to occupy 83% of stomach contents of the slenderbilled
shearwater in wet weight in the Okhotsk Sea and even higher percentages in some
part of the Bering Sea (Ogi et al., 1980). Ogi et al. (1980) suggested that the
slender-billed shearwater consume 30 000 tonnes of Thysanoessa raschi in Bristol Bay
during their stay from late April to early June assuming each bird feeds on 90 g of T.
186
Krill: Biology, Ecology and Fisheries
raschi per day. The consumption of euphausiids by the slender-billed shearwater is
almost equal to that consumed by sockeye salmon.
Although seabird observations during winter are rare in Sanriku waters, Ogi
(1994) found a large number of auklets present in late February to early March.
These include the thick-billed murre (Uria lomvia), the common murre (U.
aalge), the rhinoceros auklet, the crested auklet (Aethia cristatella), the ancient
murrelet (Synthliboramphus antiquum) and so on. These auklets seem to feed on
euphausiids, almost exclusively E. pacifica, and copepods, because dense patches
of zooplankton are always being detected on the echogram under small flocks of
crested auklets. Ogi suggests that these auklets are not endemic to Sanriku
waters, but come from the Okhotsk Sea to overwinter in the area. The importance of euphausiids as prey for thick-billed murres in the North Pacific ranks
third following squids and fish (Ogi, 1980). The largest number of euphausiids
found in a single thick-billed murre was 2246 in the East Kamchatka Current.
Quantitative estimates of E. pacifica consumed by these seabirds in Sanriku
waters are also needed.
Energy flow through Euphausia pacifica
Wada & Yabuki (1994) constructed a trophodynamic model for the continental
shelf ecosystem of the Oyashio region. They took into account 4 physical factors
Fig. 7.1.1 Structure of the trophodynamic model in the shelf area of the Oyashio region (Wada &
Yabuki, 1994).
Role of Krill in Marine Food Webs
187
(light, nutrients, temperature and mixed layer depth), 7 biological factors (diatoms,
copepods, euphausiids, sardines, walleye pollack, fisheries and diatom sinking), and
21 processes which link these factors (Fig. 7.1.1). It is interesting to note that a
deepening of the mixed layer depth and an increase in water temperature lead to
Fig. 7.1.2 Energy flow through Euphausia pacifica in the marine ecosystem of Sanriku and Joban
waters based on various sources referred in the text.
188
Krill: Biology, Ecology and Fisheries
increased euphausiid production. Conversely, a shoaling of the mixed layer depth
and a decrease in solar radiation and water temperature decrease it.
The energy flow through E. pacifica in the marine ecosystem of Sanriku and
Joban areas is summarised in Fig. 7.1.2. E. pacifica eats mainly diatoms in spring
and copepods, dinoflagellates and tintinnids in other seasons (Endo, 1981, Nakagawa et al., in preparation). Copepods, however, seem to be most important as
prey for E. pacifica throughout the year in terms of carbon (Nakagawa et al., in
preparation). In spring, E. pacifica forms dense swarms and is pelagic both day
and night. Blacktailed gull, auklets, sand lance and minke whales target the
swarms. In other seasons E. pacifica moves to offshore waters and forms benthopelagic aggregations during daytime. Therefore, many pelagic and demersal predators eat E. pacifica.
References
Akiyama, M. (1996) Body length composition of Euphausia pacifica in the stomach contents
of blue eye, Chlorophthalmus albatrossis (borealis), in Joban coastal waters. Rep. Res.
Meetings on North Pacific Krill Resources 5, 99±100 (in Japanese).
Endo, Y. (1981) Ecological Studies on the Euphausiids Occurring in the Sanriku Waters with
Special Reference to Their Life History and Aggregated Distribution. Ph.D. thesis,
Tohoku University, Sendai, 166 pp. (in Japanese with English abstract).
Fujita, T. (1994) Importance of planktonic organisms as food of demersal fish population.
Gekkan Kaiyo (Kaiyo Monthly) 26, 236±41 (in Japanese).
Fujita, T. & Ohta, S. (1989) Spatial structure within a dense bed of the brittle star Ophiura
sarsii (Ophiuroidea: Echinodermata) in the bathyal zone off Otsuchi, northeastern Japan.
J. Oceanogr. Soc. Jap. 45, 289±300.
Kasamatsu, F. & Tanaka, S. (1992) Annual changes in prey species of minke whales taken off
Japan 1948±87. Nippon Suisan Gakkaishi 58, 637±51.
Kodama, J. & Izumi, Y. (1994) Factors relevant to the fishing ground formation of Euphausia
pacifica and the relation to the demersal fish resources. Gekkan Kaiyo (Kaiyo Monthly) 26,
228±35 (in Japanese).
Komaki, Y. (1967) On the surface swarming of euphausiid crustaceans. Pacif. Sci. 21, 433±48.
Kuroda, K. (1994) Euphausiid fishery in the Japanese waters and related research activities.
Gekkan Kaiyo (Kaiyo Monthly) 26, 203±209 (in Japanese).
Minami, H., Minagawa, M. & Ogi, H. (1995) Changes in stable carbon and nitrogen isotope
ratios in sooty and short-tailed shearwaters during their northward migration. The Condor
97, 565±74.
Nakagawa, Y., Endo, Y. & Taki, K. (in press) Diet of Euphasia pacifica Hansen in Sanrika
waters off northeastern Japan. Plankton Biol. Ecol.
Ogi, H. (1980) The pelagic feeding ecology of thick-billed murres in the North Pacific,
March±June. Bull. Fac. Fish., Hokkaido Univ. 31, 50±72.
Ogi, H. (1994) Euphausiids as food of sea birds. Gekkan Kaiyo (Kaiyo Monthly) 26, 242±47
(in Japanese).
Ogi, H., Kubodera, T. & Nakamura, K. (1980) The pelagic feeding ecology of the short-tailed
Role of Krill in Marine Food Webs
189
shearwater Puffinus tenuirostris in the subarctic Pacific region. J. Yamashina Inst. Ornith.
12, 157±82.
Stancyk, S.E., Fujita, T. & Muir, C. (1998) Predation behavior on swimming organisms by
Ophiura sarsii. In Echinoderms: San Francisco (R. Mooi & M. Telford, eds), pp. 425±29.
Balkema, Rotterdam.
Takeuchi, I. (1985) A preliminary report on the stomach contents of sand lance in the coastal
area of Miyagi Prefecture. Bull. Tohoku Branch Jap. Soc. of Sci. Fish. 35, 19±21 (in
Japanese).
Wada, T. & Yabuki, K. (1994) Trophodynamic model in the shelf area of the Oyashio region.
Gekkan Kaiyo (Kaiyo Monthly) 26, 251±55 (in Japanese).
Yamamura, O. (1993) Resource partitioning among demersal fishes off Sendai Bay. GSK
Northern Japan Demersal Fish Res. Group Rep. 26, 61±70 (in Japanese).
Yamamura, O., Inada, T. & Shimazaki, K. (1998) Predation on Euphausia pacifica by
demersal fishes: predation impact and influence of physical variability. Mar. Biol. 132, 195±
208.
190
Krill: Biology, Ecology and Fisheries
7.2
Canadian waters
Ron Tanasichuk
The Canadian west, east and Arctic coasts certainly support fish, bird and mammalian species that would depend on krill as food. Most information is descriptive.
Consumption estimates exist only for pelagic fish from the south-west coast of
Vancouver Island, on the Canadian west coast.
Pacific hake (Merluccius productus), Pacific herring (Clupea pallasi) and spiny
dogfish (Squalus acanthias) dominate the pelagic fish biomass along the south-west
coast of Vancouver Island. Tanasichuk (1999a) reported that euphausiids accounted
for an average of 84 and 89% of the daily ration of hake and dogfish respectively
over August 1985±97, and for all of the daily ration for herring over those years (Fig.
7.2.1). Tanasichuk (1999b) reported that hake persist in preferring Thysanoessa
spinifera and euphausiids longer than 17 mm even though there has been a large
reduction in euphausiid biomass, a shift to smaller animals in the population and a
change in species composition following the 1992 ENSO event (Tanasichuk, 1998
a,b). Ware and McFarlane (1995) comment on how euphausiid distributions influence hake movements. Waddell et al. (1992) and Morris & Healey (1990) found that
chinook (Onchorynchus tschawytscha) and coho (O. kisutch) salmon feed heavily
on euphausiids. Other significant regions on the west coast are the Strait of Georgia
and Hecate Strait. Euphausiids are important prey for the resident hake stock,
young Pacific herring and coho, chinook and chum (O. keta) salmon in the Strait of
Georgia (R.J. Beamish, Fisheries and Oceans Canada, pers. comm.) They are also
taken when available to flatfishes in Hecate Strait (J. Fargo, Fisheries and Oceans
Canada, pers. comm.).
Perry & Waddell (1997) summarised information on zooplankton feeding by
seabirds found in west coast waters. Euphausiids were specifically identified as prey
for 17 of the 24 species. T. spinifera, T. longipes and Euphausia pacifica were
commonly observed in the diets. Euphausiids are not important food for birds in the
Strait of Georgia (Mackas & Fulton, 1989).
A number of baleen whale species feed on euphausiids along the Pacific and
Atlantic coasts every summer (G. Ellis, Fisheries and Oceans Canada, pers. comm.).
Blue (Balenoptera musculus), humpback (Megaptera novaeangliae), fin (Balenoptera physalus) and minke (Balenoptera acutorostrata) whales rely heavily on krill.
Recently, humpbacks have shifted away from the south-west coast of Vancouver
Island. This is coincident with the decline in euphausiid biomass after the 1992
ENSO event.
Euphausiids are not important prey in the Canadian Arctic. Gaston & Bradstreet
(1993) reported thick-billed murres (Uria lomvia), the most numerous nesting
seabird in the eastern Canadian Arctic, fed on hyperiid amphipods. Euphausiids
were a minor prey item for Arctic cod (Boreogadus saida) in Beaufort Sea coastal
Role of Krill in Marine Food Webs
191
Fig. 7.2.1 Daily ration estimates (% wet body mass day±1) for herring, hake and dogfish, August
1985±97. Solid lines are the euphausiid ration and dotted lines are total ration. Bars are 95%
confidence limits.
waters, Alaska (Craig et al., 1982). Bowhead whales (Balaena mysticetus) feed on
euphausiids.
Euphausiids are important prey for fish species feeding along the east coast of
Canada. They were important for silver hake (Merluccius bilinearis) and cod
(Gadus morhua Linneaus) along western Nova Scotia and haddock (Melanogrammus aeglefinus) in the Georges Bank area (Maurer, 1975). Young silver hake,
red hake (Urophycis chuss Walbaum), adult herring (Clupea harengus L.) and
mackerel on Georges Bank fed mainly on euphausiids (Meganytiphanes) (Vinogradov, 1984). The dominant fish species in the Scotian shelf area feed mainly on
euphausiids (Harding 1996). Vesin et al. (1981) found that euphausiids (T. raschii)
dominated the diet of capelin (Mallotus villosus) in the western Gulf of St. Lawrence. Pedersen & Riget (1992) found that euphausiids were important prey for
192
Krill: Biology, Ecology and Fisheries
redfish (Sebastes sp.) off west Greenland. Lilly & Rice (1983) reported that T.
raschii was an important prey item for Atlantic cod feeding on the northern Grand
Bank in spring. Euphausiids were important prey for three (beaked redfish
(Sebastes mentalla), Acadia redfish (S. fasciatus) and longfin hake (U. chesteri)) of
the 14 species collected from the Flemish Cap, about 320 km off the Newfoundland
Coast (Rodriguez-Marin et al., 1994).
There is diet information for procellariiform and alcid marine birds from the east
coast. Brown et al. (1981 cited in Prince & Morgan, 1987) found that, in seasons
when Meganyctiphanes and sandlance Ammodytes were available, Puffinus gravis
mainly took fish and P. griseus took euphausiids. M. norvegica was important food
for Leach's storm petrel (Oceanodroma leucorhoa) off Nova Scotia (Linton, 1978,
cited in Prince & Morgan, 1987). Bradstreet (1983, cited in Bradstreet & Brown,
1985) found that adult razorbill (Alca torda ), one of the five alcid species occurring
along the east coast, fed on sculpins and euphausiids.
References
Bradstreet, M.S.W. & Brown, R.G.B. (1985) Feeding ecology of the Atlantic Alcidae. In The
Atlantic Alcidae: The Evolution, Distribution and Biology of the Auks Inhabiting the
Atlantic Ocean and Adjacent Waters (D.N. Nettleship and T.R. Birkhead, eds), pp. 264±318.
Academic Press, London.
Craig, P.C., Griffiths, W.B., Haldorson, L. & McElderry, H. (1982) Ecological studies of
Arctic cod (Boreogadus saida) in Beaufort Sea coastal waters, Alaska. Can. J. Fish. Aquat.
Sci. 39, 395±406.
Gaston, A.J. & Bradstreet, M.S.W. (1993) Intercolony differences in the summer diet of
thick-billed murres in the eastern Canadian Arctic. Can. J. Zoology 71, 1831±40.
Harding, G.C.H. (1996) Ecological factors to be considered in establishing a new krill fishery
in the Maritimes region. DFO Atlantic Fisheries Research Document 96/99. 11 pp.
Lilly, G.R. & Rice, J. C. (1983) Food of Atlantic cod (Gadus morhua) on the northern Grand
Bank in spring. NAFO SCR Document 83/9/87. 35 pp.
Mackas, D.L. & Fulton, J.D. (1989) Distribution and aggregation of zooplankton in the Strait
of Georgia and their potential availability to marine birds. In The Ecology and Status of
Marine and Shoreline Birds in the Strait of Georgia, British Columbia (K. Vermeer & R.W.
Butler eds). Proceedings of a symposium sponsored by the Pacific Northwest Bird and
Mammal Society and the Canadian Wildlife Service, held in Sidney, B. C. 11 December
1987, pp. 19±25.
Maurer, R. (1975) A preliminary description of some important feeding relationships.
ICNAF Research Document 75/IX/130. 15 pp.
Morris, J.F.T. & Healey, M.C. (1990) The distribution, abundance and feeding habits of
chinook and coho salmon on the fishing banks. Canadian Technical Report of Fisheries and
Aquatic Sciences 1759, 75 pp.
Pedersen, S.A. & Riget, F. (1992) Feeding habits of redfish, Sebastes sp., in West Greenland
waters with special emphasis on predation on shrimp. ICES Document C.M. 1992/G:24.
Perry, R.I. & Waddell, B.J. (1997) Zooplankton in Queen Charlotte Island waters: distribution and availability to marine birds. In The Ecology, Status, and Conservation of
Role of Krill in Marine Food Webs
193
Marine and Shoreline Birds of the Queen Charlotte Islands (K. Vermeer & K.H. Morgan,
eds). Canadian Wildlife Service. Occasional Paper 93, pp. 18±28.
Prince, P.A. & Morgan, R.A. (1987) Diet and feeding ecology of Procellariiformes. In Seabirds: Feeding Ecology and Role in Marine Ecosystems (J.P. Croxall, ed.), pp. 154±71.
Cambridge University Press, Cambridge.
Rodriguez-Marin, E., Punzon, A., Paz, J. & Olaso, I. (1994) Feeding of most abundant fish
species in Flemish Cap in summer 1993. NAFO SCR Document. 94/35. 33 pp.
Tanasichuk, R.W. (1998a) Interannual variations in the population biology and productivity
of the euphausiid Thysanoessa spinifera in Barkley Sound, Canada, with special reference
to the 1992 and 1993 warm ocean years. Mar. Ecol. Prog. Ser. 173, 163±80.
Tanasichuk, R.W. (1998b) Interannual variations in the population biology and productivity
of the euphausiid Euphausia pacifica in Barkley Sound, Canada, with special reference to
the 1992 and 1993 warm ocean years. Mar. Ecol. Prog. Ser. 173, 181±95.
Tanasichuk, R.W. (1999a) Euphausiids in a coastal upwelling ecosystem: Their importance as
fish prey, interannual variations in euphausiid population biology and productivity, and
some implications of these changes for fish production. PhD. thesis, University of Bergen,
Norway. 123 pp.
Tanasichuk, R.W. (1999b) Interannual variations in the availability and utilization of
euphausiids as prey for Pacific hake (Merluccius productus) along the south-west coast of
Vancouver Island. Fish. Oceanogr. 8, 150±56.
Vesin, J.-P., Leggett, W.C. & Able, K.W. (1981) Feeding ecology of capelin (Mallotus villosus) in the estuary and western Gulf of St. Lawrence and its multispecies implications.
Can. J. Fish. Aquat. Sci. 38, 257±67.
Vinogradov, V.I. (1984) Food of silver hake, red hake and other fishes of Georges Bank and
adjacent waters, 1968±74. Science Council Studies NAFO 7, 87±94.
Waddell, B.J., Morris, J.F.T. & Healey, M.C. (1992) The abundance, distribution, and biological characteristics of chinook and coho salmon on the fishing banks off southwest
Vancouver Island, May 18±30, 1989 and April 23±May 5, 1990. Can. Tech. Rep. Fish. Aquat.
Sci. 1891, 113 pp.
Ware, D.M. & McFarlane, G.A. (1995) Climate-induced changes in Pacific hake (Merluccius
productus) abundance and pelagic community interactions in the Vancouver Island
Upwelling System. In Climate change and northern fish populations (R.J. Beamish ed.),
pp. 509±521. Canadian Special Publication of Fisheries and Aquatic Sciences 121, 509±21.
194
Krill: Biology, Ecology and Fisheries
7.3
The Southern Ocean
Inigo Everson
The Southern Ocean food web
Generalised descriptions of the Southern Ocean food web have been given by
several workers with good examples to be found in Hart (1942), Holdgate (1967)
and Murphy et al (1988). These, by grouping several related species together and
presenting them as single components, highlight what is perceived as the relative
simplicity of the system. An oft-quoted example is the food chain which leads from
phytoplankton, around a millimetre in size, through Antarctic krill, which grow to
around six centimetres, to the blue whale which can grow up to 30 metres in length.
Simplicity is only a relative term as even though the system may be simple relative to
other oceanic ecosystems, when one examines each component, it is clear that the
picture in Fig. 7.3.1 is in reality quite complicated.
The Convention for the Conservation of Antarctic Marine Living Resources
(CCAMLR) provides a slightly different focus, concentrating on two categories, the
Fig. 7.3.1 Simplified representation of Southern Ocean Food Web linkages that are centred
around krill.
Role of Krill in Marine Food Webs
195
harvested and dependent species. If the list is restricted to those species for which
there is currently a targeted fishery, then in the Southern Ocean, there are relatively
few that fall into the category of harvested species. Apart from the Antarctic krill
(Euphausia superba), there is the mackerel icefish (Champsocephalus gunnari) and
Patagonian and Antarctic toothfish (Dissostichus eleginoides and D. mawsoni
respectively). There have been in addition exploratory fisheries for stone crab
(Paralomis spp) and squid (Martialia hyadesi). Other species have been harvested in
the past and information on the reported catches can be found in the Statistical
Bulletins (CCAMLR, 1999). Of these species the only one which is itself harvested
and is at the same time dependent on krill is the mackerel icefish.
Krill fishing, which in the Southern Ocean in the 1960s was very much in embryonic
form, was considered by the 1970s to be an opportunity to provide large amounts of
protein to satisfy world demands. As the krill fishery expanded there was concern that
it might adversely affect dependent species. This heightened awareness was behind
the research programme `Biological Investigations of Marine Antarctic Systems and
Stocks' (BIOMASS) (SCAR/SCOR, 1977). This was the research programme which
in a short space of time led to the Commission for the Conservation of Antarctic
Marine Living Resources (CCAMLR) (Miller & Agnew in Chapter 12). Central to
both is a requirement for the identification of the key dependent species and the
quantification of interactions between them and krill. The dependent species for
which there is greatest information published are the whales, seals, birds and fish.
Impact of major dependent species
Reviewing the diet of Southern Ocean birds, Croxall (1984) identified the following
species that are dependent to a great extent on krill. Of the penguins, Adelie
(Pygoscelis adelie), chinstrap (P. antarctica), macaroni (Eudyptes chrysolophus) and
gentoo (P. papua) feed extensively on Antarctic krill (Euphausia superba), although
in high latitudes, such as in Prydz Bay, Adelies feed on E. crystallorophias (Cooper
& Woehler, 1994). In the northern part of the Southern Ocean around South
Georgia both gentoo and macaroni penguins also feed on E. frigida and in the same
region Macaroni penguins also feed on E. triacantha.
Although black-browed (Diomedea melanophris), light-mantled sooty (Phoebretia palpebrata) and grey-headed (D. chrysostoma) albatrosses feed on krill this
constitutes only 40% of the diet for the first two and 16% for grey-heads (Croxall,
1984; Prince & Morgan, 1987). Only about one-fifth of the diet of giant petrels
(Macronectes spp) contains krill although, as with many of the Procellariidae, their
diet appears to be fairly catholic. Some of the smaller petrels, such as the Antarctic
petrel (Thalassoica antarctica), Cape petrel (Daption capensis), snow petrel
(Pagodroma nivea) and diving petrel (Pelecanoides spp) also feed extensively on
krill. The smaller storm petrels (Oceanites spp) and prions (Pachyptillia spp) feed on
a wider range of crustacea including krill although the emphasis is much more
towards copepods (Prince & Morgan, 1987).
196
Krill: Biology, Ecology and Fisheries
That the birds are significant consumers of krill can be seen from the estimates of
the annual amount of krill eaten by the different species. Considering two localities,
South Georgia at the northern end of the zone and Prydz Bay close to the Antarctic
continent, we can see in Table 7.3.1. the differences in the amounts taken. This also
highlights the fact that the bulk of the consumption of krill by birds is attributable to
relatively few species, macaroni penguins and prions at South Georgia and Adelie
penguins at Prydz Bay.
Table 7.3.1 Estimated annual consumption in thousands of tonnes of krill by major bird species.
The figures in parenthesis are the proportion by mass that krill represents of the diet of the species
at that site. Data from Croxall & Prince (1987) for South Georgia and Cooper & Woehler (1994)
for Prydz Bay.
Species
Adelie penguin
Chinstrap penguin
Gentoo penguin
Macaroni penguin
Black-browed albatross
Grey-headed albatross
Light mantled sooty albatross
Southern giant petrel
Northern giant petrel
Antarctic petrel
Cape petrel
Antarctic fulmar
Snow petrel
Antarctic prion
White-chinned petrel
Wilson's petrel
Common diving petrel
South Georgia diving petrel
South Georgia
±
2.5 (100)
61
(68)
3872
(98)
8.3 (38)
6.3 (15)
1.8 (37)
2.7 (82)
1.5 (65)
±
3.1 (86)
±
0.4 (80)
1345
(58)
210
(27)
3.8 (40)
52.8 (15)
126
(76)
Prydz Bay
66
0
0.53
0.14
0.98
0.01
(97)
±
±
±
±
±
±
±
±
±
(0.77)
(0.21)
(1.44)
(0.02)
0.23 (0.33)
±
±
With the exception of elephant seal (Mirounga leonina) all the Antarctic seals
feed to some extent on krill (Laws, 1984). In the case of Weddell (Leptonychotes
weddelli) and Ross (Ommatophoca rossi) the proportion of krill in the diet is
relatively slight. Krill are the major item in the diet of crabeater seals (Lobodon
carcinophagus) and make up around half the food eaten by leopard seals (Hydrurga
leptonyx). Fur seals (Arctocephalus gazella) feed extensively on krill but when that
is scarce concentrate on fish and squid. This change of diet has been implicated in
the large changes in standing stock of mackerel icefish around South Georgia
(Everson et al., 1999). The diving depths of fur and crabeater seal have been shown
to follow the diurnal pattern of vertical migration of krill (Croxall et al., 1985; Laws,
1984). This indicates that although krill may be present in an area, if they remain
deep, they may be out of the diving range of the seals. The latitudinal range of
foraging activity of seals is shown in Fig. 7.3.2.
By far the most abundant seal species in the Southern Ocean is the crabeater seal
Role of Krill in Marine Food Webs
197
Fig. 7.3.2 Latitudinal distribution of foraging ranges of whales and seals that are dependent on
krill in the Southern Ocean. The checked area is the main range of foraging activity. With permission from The Royal Society, London. (Modified from Laws, 1977.)
the total population of which in 1956±57 was thought to number about 15 million
individuals (Eklund & Atwood, 1962) and 30 million in 1982 (Laws, 1984). This
estimate was further revised by Erickson & Hanson (1990) who gave a population
size of 11 to 12 million individuals. The differences between these estimates may
represent some underlying changes although they are more likely to reflect the
difficulties in counting seals which are concentrated in the pack ice zone. Their
distribution within the pack ice zone means that their feeding area will be changing
with the seasonal changes in sea-ice distribution; thus in winter they may be present
much further to the north than during the summer when the ice cover is at a
minimum. Taking the consumption rates given by Laws (1984) this indicates that
crabeater seals consume between 50 and 134 million tonnes of krill a year. Although
fur seals are important locally around South Georgia where the current population
of around 4 million (Boyd, 1993) is estimated to take around 5 or 6 million tonnes of
krill per annum (Everson et al., 1999), their impact on a Southern Ocean scale
remains quite small.
Seven species of baleen whale are present in the Southern Ocean, all of which have
been subject to extensive whaling. The latitudinal range of foraging activity by the
five species (Minke Balaenoptera acutorostrata, Blue B. musculus, Fin B. physalus,
Sei B. borealis and humpback Megaptera novaeangliae) which feed predominantly on
krill is shown in Fig. 7.3.2. Estimates of stock sizes and their estimated consumption of
krill prior to the onset of commercial whaling and from recent whale sighting surveys
are set out in Table 7.3.2. Blue, fin and sei whales migrate into the Southern Ocean
during the summer to feed (Plate 4, facing p. 182). The southerly limit of distribution
on these feeding migrations is the northern limit of the pack ice zone so that their
198
Krill: Biology, Ecology and Fisheries
Table 7.3.2 Stock sizes and estimated krill consumption of initial and present whale stocks present in the Southern Ocean. Stock sizes from Kock & Shimadzu (1994), krill consumption rates
derived from Laws (1977).
Whale
species
Initial stock
size
(thousands)
Present
stock size
(thousands)
220
490
200
130
?
0.7±5.5
25
38
12
760
Blue and pygmy blue
Fin
Sei
Humpback
Minke
TOTAL
Initial whale
stock krill
consumption
(million tonnes)
Present whale
stock krill
consumption
(million tonnes)
80
100
15
14
?
> 209
0.24±1.9
5
2.7
1.3
75
~85
foraging region changes as the pack ice retreats. Results of an analysis of whale
distributions and pack ice limits over the period 1931±1939 presented by Mackintosh
(1970) are shown in Fig. 7.3.3. In a study on Minke whales in the Ross Sea and the area
immediately to the west, Ichii et al. (1998) found that the amount of body fat was
related to the feeding conditions during that season. This was related further to the
amount and size of krill and also compared to the extent of sea-ice cover.
0
5
Oct.
Nov.
Dec.
Jan.
Feb.
Mar.
13.7
14.7
16.8
19.6
22.1
24.9
Antc. conv.
km 2/10 6
10
15
2.1
3.0
4.2
20
Total zone
grazed
16.5 × 10 6 km 2
Fee
25
30
ding
zon
e
7.3
18.2
Pack-ice
7.3
5.1
16.3
13.0
6.2
3.2
Total area (A.C. – Shelf): 34 × 10 6 km 2
2.4
Contl.
shelf
Fig. 7.3.3 Seasonal variation in the ice-free area between the Antarctic Polar Frontal Zone and
the Antarctic Continent and indication of the potential grazing area for blue and fin whales. (With
permission from Mackintosh, 1970.)
Role of Krill in Marine Food Webs
199
Several species of fish are known to feed on krill (Kock, 1992) and the concentrations of some species, such as mackerel icefish (Champsocephalus gunnari) at
South Georgia, or Myctophidae in the open sea may be locally significant. There is
evidence for local availability affecting the fish populations (Everson et al., 1999)
and this topic is developed further in Everson, Chapter 8. In spite of this the overall
consumption of krill by fish on a Southern Ocean scale is unlikely to be anything like
as great as that of whales, seals or birds. Squid are a group known to be present in
the Southern Ocean some of which are known to feed on krill. Kock & Shimadzu
(1994) discussed this at some length but concluded that, beyond noting that the
amount might be very large and probably at least tens of millions of tonnes, precise
figures were impossible to determine.
The foregoing discussion indicates two important points, firstly that relatively few
species have a very large impact on krill stocks. Secondly, that if the estimated
amount of krill consumed by predators is used as the indicator, the annual production
of krill must be of the order of hundreds of millions of tonnes. Even if these consumption figures are overestimated by a substantial margin the total is still well above
the reported krill catch for the Southern Ocean (Miller & Agnew in Chapter 12).
Ecosystem monitoring
The static picture of a food web providing qualitative links between the different
elements is only the starting point for understanding ecosystem interactions. An
obvious development is to monitor key components and assess how they themselves
vary and in relation to other parts of the system. Recognising the central role of krill
in the Southern Ocean system, CCAMLR established its own Ecosystem Monitoring Programme (CEMP). This arose from a meeting where the objective of
ecosystem monitoring was defined as:
To detect and record significant changes in critical components of the ecosystem,
to serve as a basis for the Conservation of Antarctic Marine Living Resources.
The monitoring system should be designed to distinguish between changes due to
the harvesting of commercial species and changes due to environmental variability, both physical and biological.
(SC-CAMLR 1985)
Discussion, at that meeting and subsequently, centred around programmes that
were then in operation and which might, with minor adjustment, be modified to
fulfil the aims of the CEMP and additional studies which would need to be set de
novo. Arising from this a series of key species comprising penguins, flying birds and
seals was identified for which information could be collected routinely in a standardised manner so as to include the main components of the life history. These
Standard Methods are set out in SC-CAMLR (1991). Analysing these data involves
an assessment of the spatial and temporal scales of the interactions, a subject that is
considered in more detail in Everson (Chapter 8).
200
Krill: Biology, Ecology and Fisheries
References
Boyd, I.L. (1993) Pup production and distribution of breeding Antarctic fur seals (Arctocephalus gazella) at South Georgia. Antarctic Sci. 5, 17±24.
CCAMLR (1999) Statistical Bulletin. Commission for the Conservation of Antarctic Marine
Living Resources, Hobart, Australia.
Cooper, J. & Woehler, E.J. (1994). Consumption of Antarctic krill (Euphausia superba) by
seabirds during summer in the Prydz Bay region, Antarctica. In Southern Ocean Ecology:
the BIOMASS Perspective (S. El-Sayed, ed.), pp. 247±60. Cambridge University Press,
Cambridge.
Croxall, J.P. (1984) Seabirds. In Antarctic Ecology (R. Laws, ed.), pp. 531±616. Cambridge
University Press, Cambridge.
Croxall, J.P. & Prince, P.A. (1987) Seabirds as predators on marine resources, especially krill,
at South Georgia. In Seabirds. Feeding Ecology and Role in Marine Ecosystems (J.P.
Croxall, ed.), pp. 347±68. Cambridge University Press, Cambridge.
Croxall, J.P., Everson, I., Kooyman, G.L., Ricketts, C. & Davis, R.W. (1985) Fur seal diving
behaviour in relation to vertical distribution of krill. J. Anim. Ecol. 54, 1±8.
Eklund, C.R. & Atwood, E.L. (1962) A population study of Antarctic Seals. J. Mammalogy
43, 229±38.
Erickson, A.W. & Hanson, M.B. (1990) Continental estimates and populations trends of
Antarctic ice seals. In: Antarctic Ecosystems. Ecological Change and Conversation (K.R.
Kerry & G. Hempel, eds), pp. 253±64. Springer-Verlag, Berlin.
Everson, I., Parkes, G., Kock, K.-H. & Boyd, I. (1999) Variation in the standing stock of the
mackerel icefish Champsocephalus gunnari at South Georgia. J. Appl. Ecol. 36, 591±603.
Hart T.J. (1942) Phytoplankton periodicity in Antarctic surface waters. Discovery Rep. 21,
261±356.
Holdgate, M.W. (1967) The Antarctic Ecosystem. Phil. Trans. R. Soc. 252, 363±89.
Ichii, T., Shinohara, N., Fujise Y., Nishiwaki, S. & Matsuoka, K. (1998) Interannual changes in
body fat condition index of minke whales in the Antarctic. Mar. Ecol. Prog. Ser. 175, 1±12.
Kock, K.-H. (1992) Antarctic Fish and Fisheries. Cambridge University Press, Cambridge.
Kock, K.-H. & Shimadzu, Y. (1994) Trophic relationships and trends in population size and
reproductive parameters in Antarctic high-level predators. In Southern Ocean Ecology: the
BIOMASS perspective (S. El-Sayed, ed.), pp. 287±312. Cambridge University Press,
Cambridge.
Laws, R.M. (1977) Seals and whales in the Southern Ocean. In Scientific Research in Antarctica. Discussion meeting organised by V.E. Fuchs & R.M. Laws. Phil. Trans. Roy. Soc.
London B279 81±96.
Laws, R.M. (1984) Seals. In: Antarctic Ecology (R. Laws, ed.), pp. 531±616. Cambridge
University Press, Cambridge.
Mackintosh, N.A. (1970) Whales and krill in the twentieth century. In Antarctic Ecology, Vol.
1 (M.W. Holdgate, ed.), pp. 195±212. Academic Press, London.
Murphy, E.J., Morris,D.J., Watkins, J.L. & Priddle, J.L. (1988) Scales of interaction between
Antarctic krill and the environment. In Antarctic Ocean and Resources Variability (D.
Sahrhage, ed.), pp. 120±30. Springer-Verlag, Berlin.
Prince P.A. & Morgan, R.A. (1987) Diet and feeding ecology of Procellariiformes. In Seabirds. Feeding Ecology and Role in Marine Ecosystems (J.P. Croxall, ed.), pp. 154±71.
Cambridge University Press, Cambridge.
Role of Krill in Marine Food Webs
201
SCAR/SCOR (1977) Biological Investigations of Marine Antarctic Systems and Stocks
(BIOMASS). Vol 1: Research Proposals, 79 pp. University Library, Cambridge.
SC-CAMLR (1985) Report of the ad hoc Working Group on Ecosystem Monitoring, Report
of the Fourth Meeting of the Scientific Committee for the Conservation of Antarctic
Marine Living Resources. CCAMLR, Hobart Australia.
SC-CAMLR (1991) CCAMLR Ecosystem Monitoring Program (CEMP). Standard Methods
for Monitoring Studies, 131 pp. CCAMLR, Hobart, Australia.
Chapter 8
Ecosystem Dynamics Involving Krill
Inigo Everson
8.1
Introduction
The picture of krill that has been developed in the preceding chapters will have
demonstrated the central role of krill in oceanic food webs. Recognition of this, and
the potential effects that overfishing on krill might have on dependent species, was
behind the requirement for an ecosystem approach to management enshrined in the
Convention for the Conservation of Antarctic Marine Living Resources
(CCAMLR). This has in turn focused attention on the provision of scientific advice
and for the ecosystem approach this is much more complicated than the traditional
`single-species' assessments.
Having accepted the ecosystem approach, the supporting scientific advice needs
an increased level of complexity. This is because we need to determine not only the
direct effects of harvesting on the resource but also whether a given level of fishing
will adversely impact dependent species. This is further complicated by the need to
take account of natural variation in the system. Understanding these sources of
variation, whether they be associated with the krill, the fishery, the environment,
dependent species or interactions between them, is a topic which requires a careful
examination of the available information leading to the development of models
which seek to provide explanations.
Traditional single-species fisheries models use the population as the basic unit for
which a total allowable catch (TAC) is determined. This is useful in providing an
upper limit beyond which the target fishery should not expand although it should
not be used in isolation. When this first requirement has been met, further measures
can be put in place to take account of the spatial and temporal scales of the interactions between the ecosystem components. This immediately raises a further level
of complexity beyond single species population models because there is an enormous number of possible interactions within the system which are themselves
operating at different time and space scales. In this chapter I follow a line starting
with the simplest approach at the large scale and develop this through the time and
space scales appropriate for the interactions leading to a consideration of ecosystem
assessment and how this can be used to provide advice for management of krill
fisheries.
202
Ecosystem Dynamics Involving Krill
8.2
203
Large-scale interactions
At the large scale we are looking for models which will provide an umbrella value
with which to define an absolute maximum for the fishery. One of the simplest
models of this type is that defined by Gulland (1971) who was looking for a simple
approximation for maximum sustainable yield (MSY) in situations where data are
sparse. His basic formulation:
Y ˆ 0:5MB0
where M is the coefficient of natural mortality and B0 is the unexploited biomass,
was further developed by Beddington and Cooke (1983). They demonstrated that
the factor 0.5 was generally too high due to uncertainties over the estimation of M
and recruitment (R) and that this would result in many instances in MSY being
overestimated.
Their formulation, Y = lMB0, was used by Butterworth et al. (1992, 1994) to
provide a method for calculating the potential yield of krill. This Krill Yield Model
(KYM) was then developed further as the Generalised Yield Model (GYM) by
Constable & de la Mare (1996) and which incorporated the terms l and M into a
single constant g to give a basic formulation: Y = gB0. The derivation and characteristics of g are described by Miller & Agnew in Chapter 12. Thus with an M of
0.6 the original Gulland formula would have given an MSY of 0.3 times the
unexploited biomass; the value of g used for the recent estimates, as explained by
Miller & Agnew, Chapter 12, is 0.116. The CCAMLR approach is therefore erring
very much towards protecting the stock and taking account of uncertainty.
Within the CCAMLR area the precautionary catch limits derived in this way
apply to very large tracts of ocean, several million square kilometres. These
umbrella values are deemed `precautionary' because they allow for dependent
species. If all the catch were taken within a limited region, and if that same region
was critical for a dependent species, then, even though the catch limit allows for the
requirements of the dependent species in toto they may, even so, be adversely
affected. To cater for this eventuality we need further models at smaller scales.
8.3
Smaller-scale interactions
Development of models at smaller scales is a tricky business requiring knowledge of
those dependent species which should be considered and their spatial and temporal
foraging behaviour. The various sections of Chapter 7 have identified the species or
groups that should be considered, so here I generalise to highlight the scales of
interest. Spatially we need to consider both horizontal and vertical planes. Horizontally we have the geographical scale while vertically we have the location in the
water column. Temporally the most important scales are probably season and time
204
Krill: Biology, Ecology and Fisheries
of day although, depending on the topic of interest, months and long-term trends are
also important.
Whales are present in all the main oceans and a large proportion of the balaenopterids feed on krill. Their large size and high mobility has meant that they
have often been considered to have the ability to move between feeding areas.
Thus if food is scarce in one region they might simply move to a new feeding
area. Consideration of the actual feeding grounds would indicate that the continental shelf or shelf break, regions of krill aggregation, would define the geographical scale while vertically they might have a diving depth of perhaps 500 m.
Seasonally they tend to migrate to the feeding grounds during the summer. Thus
although able to feed over a wide area their main impact is probably local to the
shelf and shelf break.
Pinnipeds are thought to be of major importance in the Southern Ocean and are
typified by two contrasting species. The crabeater seal (Lobodon carcinophagus) is
widespread throughout the pack ice zone throughout the year. Diurnally, the
number of seals visible on the ice, the inverse of those diving at any time, increases
when the krill are at their deepest (Laws, 1984). This optimises the diving depth over
the top 60 metres of the water column. In contrast adult fur seals (Arctocephalus
gazella) during the breeding season are restricted to the coastal region, probably
within 150 kilometres of their breeding site (Boyd, pers. comm. reported in Everson
& de la Mare, 1996). During the winter they are able to travel far although, if they
remain in the Southern Ocean feeding on krill, they may be concentrated on the
shelf or at the shelf break. Vertically most dives are restricted to little more than
60 m (Boyd & Croxall, 1992; Boyd et al., 1997).
Birds fall into two broad categories, those that feed relatively close to the surface
and those that are able to dive. Black-browed albatross (Diomedea melanophris), in
the Southern Ocean, and petrels in all oceans have feeding ranges which cover large
areas, even during the breeding season. Their ability to catch krill is constrained by
the depths to which they can dive, a limitation of around 5 or at maximum 10 m. The
other functional group of flying birds is the diving petrels and alcids, which can cover
moderate distances rapidly, are constrained by range from breeding sites, and
vertically restricted to around the top 30 m of the water column (Bradstreet &
Brown, 1986). Penguins, a functional group of birds that cannot fly, have a restricted
foraging range during the breeding season and dive to depths over 50 m (Croxall &
Lishman, 1987), somewhat deeper than the alcids.
Fish are a group which are known to feed extensively on krill in North Pacific,
North Atlantic and to a lesser extent in Southern Ocean waters. The main interactions involving fish and krill identified in Chapter 7 appear to occur on the continental shelf and shelf break. Variation in feeding intensity is probably a function of
the distribution of both the krill and the fish, as for example krill being carried onto
the shelf by the coastal current off the Japanese coast (Endo, Chapter 7). In the
vertical plane, fish are likely to be able to feed over the whole water column on the
continental shelf, some species, such as the mackerel icefish (Champsocephalus
Ecosystem Dynamics Involving Krill
205
gunnari) around South Georgia, migrating off the bottom at night to feed (Kock,
1992).
The importance of understanding the time and space scales of the predator
foraging activity was highlighted by Everson (1983) who considered the foraging
activity of two penguin species during the breeding season. The example that was
used there was of macaroni and gentoo penguins, breeding on an isolated island at
the critical time of year when the chicks are being fed. Their maximum foraging
ranges at that time are known to be quite different although both species are to a
great extent dependent on krill. The aggregated nature of the krill distribution
means that krill may not be available to both species at the same time; this is
indicated in Fig. 8.1.
Fig. 8.1 Diagrammatic representation of a krill patch relative to the foraging range of gentoo and
macaroni penguins at the height of the breeding season. The left-hand map shows an aggregation
within the range of both species whilst the aggregation in the right-hand map would only be
available to macaroni. (Redrawn from Everson, 1983.)
The vertical distribution of krill is complicated, as described in Watkins (Chapter
4) with no simple pattern emerging to describe all situations. The general pattern is
that krill tend to be deep by day while after dark they tend to be at or near the
surface. They are also present at the surface by day in some instances (see Endo,
Chapter 3). These rhythms mean that krill may only be available to surface feeding
predators for a brief period as demonstrated in Fig. 8.2.
206
Krill: Biology, Ecology and Fisheries
Sea surface
Albatross and Petrels
Seals and
Penguins
Whales
KRILL
aggregations
Fish and Squid
Seabed
Day
Night
Day
Fig. 8.2 Representation of diurnal krill availability to different types of predator. (Redrawn from
Everson, 1983.)
8.4
Dependent species indices
In order to develop interaction models it is necessary to obtain time-series of data
on the key characteristics of the dependent species. Unfortunately this does not
appear to have been undertaken as a routine in many areas. An exception is the
Southern Ocean where the CCAMLR Ecosystem Monitoring Programme (CEMP)
has set up regular monitoring of what are thought to be key parameters associated
with several krill eating species. A full description of the programme is beyond the
scope of the present chapter so what follows is a very brief summary. The dependent
species which have been identified for the CEMP are: Adelie (Pygoscelis adelie),
chinstrap (P. antarctica) and gentoo (P. papua) penguins, black-browed albatross
(Diomedea melanophris), Antarctic (Thalassoica antarctica) and Cape petrels
(Daption capensis), fur (Arctocephalus gazella) and crabeater seals (Lobodon carcinophagus). In addition krill as a harvested species and major prey item and certain
key environmental variables are also included. Most of these parameters have been
monitored as part of the programme since 1989 or 1990 although in some instances
national programmes were in place over a decade prior to that time. The CEMP
Ecosystem Dynamics Involving Krill
207
database consequently forms a very powerful archive with which to study ecosystem
interactions.
Initially these indices were considered separately and in many instances provided
indications of what have become known as good and poor krill seasons. Two major
difficulties have emerged with interpreting these data. The first is that the magnitude and diversity of the data is so great that it is impossible to view them all in a
single table and gain a reasonable overview. The second, probably more serious
problem is that the response of each index to the krill status has to be assessed
carefully with respect to the other indices. For example when looking at penguin
breeding success (the number of chicks fledged) this indicator may go down in a
poor krill season whereas the fledging weight of the chicks may be normal; this
might be because only the well-fed chicks survive. A fuller consideration of these
interactions and their attendant problems can be found in the reports of the SCCAMLR Working Group on Ecosystem Monitoring and Management.
Refining these indices into a form whereby they can be interpreted speedily has
proved quite a difficult task. Currently groups of indices which integrate krill
availability over approximately similar time and space scales are being amalgamated into combined standardised indices (CSI). This process involves estimating
the standardised normal deviate for each variable, for each species at each site and
for each season. Currently the amalgamation is in two forms using parameters in the
first instance that integrate over the summer and secondly those indices appropriate
to the winter months. These are then combined to form a time-series for each site,
an example of which is shown in Fig. 8.3.
While having considerable merit in providing a pictorial view of the situation
and highlighting the `poor krill season' the approach fails to take account of
interactions between parameters or provide any scaling factors which might take
account of their relative influence. The other problem is that it does not take
account of the time taken for any change in krill availability being noted in the
CEMP index. Therefore to gain a better understanding of the way in which the
system operates we need to take into account the extent and timing of the
responses. In essence this means recognising the onset of a poor krill period and
knowing the delay before this effect is recognisable in the predator index. This
requires an understanding of the functional relationship between the predator
and its prey.
8.5
Functional relationships
In the ecosystem context we are looking to understand the functional relationships
between predator performance and krill availability, the latter expressed in terms of
abundance and distribution. In order to achieve this we need to understand the time
and space scales of the interactions, the underlying theme of this chapter. This was
considered in some detail by CCAMLR during a meeting in Vina del Mar, Chile
208
Krill: Biology, Ecology and Fisheries
Fig. 8.3 Combined Standardised Index for CEMP parameters at South Georgia. The two arrows
indicate seasons of very low krill availability to the predators. (Data from SC-CAMLR XVI, p. 227).
(SC-CAMLR, 1992) and during that meeting questions of functional relationships,
krill availability and natural effects were discussed.
At the largest scale, Butterworth & Thomson (1995) considered the possible
effects of different levels of krill harvesting on several dependent species. This work
is discussed by Miller & Agnew in Chapter 12 and will not be referenced further
here beyond noting that considerable work will be required before the method can
be taken as the basis for scientific advice for management of the krill fishery.
Viewed on the local scale of the continental shelf, fish provide a convenient
example for study as they are present in all the major krill zones considered in this
book and in many instances have been harvested. Here I take the example of the
interactions between the mackerel icefish (Champsocephalus gunnari) and its main
food krill at South Georgia. Biological information, length, mass, sex, maturity state
and stomach contents, are available from a series of trawl surveys in the region.
Similar data have also been obtained from the commercial fisheries that took place
in the 1970s and 1980s. Using these data several indices have been calculated.
Analysis of the stomach contents of the fish from different locations on the South
Georgia shelf is quite instructive. Results from the January±February period in
Ecosystem Dynamics Involving Krill
209
three seasons are presented in Kock et al. (1994) which indicate that there is a large
degree of variation both within and between seasons in terms of stomach fullness
and stomach contents. In 1985 and 1992 no clear pattern emerges but in 1991 there
was a much larger proportion of empty stomachs and furthermore krill were only
present in the stomachs of fish caught on the eastern and northeastern portion of the
shelf. Evidence from other sources, referenced in Everson et al. (1999) indicates that
1991 and 1994 were seasons when krill were scarce and that over much of the area
mackerel icefish were feeding on the hyperiid amphipod Themisto gaudichaudii.
These observations are consistent with the known patchy distribution of krill in the
region.
Condition indices, the ratio of measured mass to the estimated mass from a
generalised length to mass relationship, varies around unity. When the index is
greater than one the fish are assumed to have been feeding well and increasing their
overall body mass. When food is in short supply body reserves will be mobilised
leading to a reduction in mass and thus a reduction in condition index. This relationship is demonstrated in Fig. 8.4 where the krill density estimates are plotted
against the total mass condition index for mature fish from research surveys
undertaken during the same month. During this study two condition indices have
been used, one using total mass and the other using gutted mass. These show
essentially the same story although the total mass index varies more widely due to
feeding and reproductive status. This difference is more accentuated when the
condition index is greater than one as there will be variation due to stomach fullness
and gonad size. When it is less than one the stomach is more likely to be empty and
the fish are less likely to come into spawning condition.
Condition index (total mass)
1.15
1.10
1.05
1.00
0.95
0.90
0.85
0
10
20
30
40
50
60
70
80
90
100
Krill density
Fig. 8.4 Condition index of mackerel icefish plotted against krill density from acoustic surveys
undertaken during the same month.
210
Krill: Biology, Ecology and Fisheries
Although the fit to the data is good it could be improved by two refinements. The
first improvement would be to ensure that the acoustic survey to estimate the
abundance of krill is over the same area as the fish are present and likely to be
feeding. The second is a refinement which relates to the response time of the fish to
the availability of their food. A fish in poor condition is likely to feed actively when
krill become available. This will lead to an immediate and rapid response in stomach
fullness, which might be 5% of the total mass when the fish is replete, thus having an
immediate effect on the condition index estimated from total mass. After a brief
period, food will be converted to muscle so that the gutted mass condition index is
likely to follow the total mass condition index but with a short time lag. The opposite
sequence would be expected when the situation changed from one of high to low
food availability. A diagrammatic representation of the time difference between the
date of observation and the time period over which the index is integrating is
indicated in Fig. 8.5. In the absence of direct observations on condition indices and
feeding status these effects have been inferred from the work of Love (1988) who
studied the effects of feeding condition and starvation on fish.
Fig. 8.5 Conceptual framework indicating the possible response times of three mackerel icefish
indices. Diet = stomach contents or stomach fullness; Ct = condition index using total mass; Cg =
condition index using gutted mass.
Feeding status is thought also to affect the reproductive cycle of the icefish.
Periods of krill scarcity prior to the onset of the gonad maturation process have been
shown to delay the start of the process (Everson et al., 2000). In extreme cases this
delay may be for three or four months but, except in extreme years, it does not
appear to prevent the fish spawning completely. Diet, condition and spawning status
all provide some indication of the feeding conditions encountered by the icefish.
The diet, stomach fullness and composition of the contents, relate to conditions in
the immediate vicinity of the fish at the time of capture. Condition takes a little time
to work through the system although since this index correlates with the acoustic
Ecosystem Dynamics Involving Krill
211
estimates of krill the response is recognisable within a month (Fig. 8.4). The
reproduction index appears to be influenced by the krill status such that when krill
are scarce the maturation process may be delayed until conditions improve.
The conceptual framework that has been developed for mackerel icefish can, with
some modification be applied to other dependent species, a process which would
have more utility than the simple combination of indices into CSI in the CEMP. To
achieve this we need to consider the proximity in time of the observation to local
krill availability, the integration period of the index and its response time to changes
in krill availability.
In the case of penguins, at the shortest time and smallest space scales, we have, as
with the fish, stomach contents and fullness during the breeding season. The analogue from the CEMP penguin indices is diet. We also have from the CEMP
foraging trip duration (CEMP Index A5), an index that uses the time taken for the
adult penguin to obtain sufficient food for itself and its chick; the longer the trip, the
lower the availability of krill. These indices have a very short response time and all
provide spot indications of the availability of krill. They are not sufficient to provide
integration over a period of anything other than, at the most, a few days and over an
area local to the breeding colony.
Krill availability over periods of a few weeks to a few months is tracked, as we
have seen above, by the condition index of the fish. This index is available for any
time of year during which sampling takes place, be it commercial fishing or research
survey. The more that this sampling is spread out over the geographical range of the
fish and throughout the year the clearer the picture that will emerge of the availability of krill.
The analogues for the penguins are much more closely specified to time of year.
Thus the weight of the adult birds on arrival at the breeding colony (CEMP Index
A1) is integrating over around four or five months prior to a set calender date. This
arrival weight may be set by a ramp function of `increasing importance' as that date
approaches which, in the case of penguins, means that it is probably related to local
krill availability, although at this stage local is not defined. At South Georgia, where
the krill fishery is concentrated in the winter months, this may be the period of
greatest predator±fishing overlap, a topic considered later. The duration of the first
incubation shift (CEMP Index A2) may be influenced not only by the arrival weight
(CEMP Index A1) but also by the availability of krill to the female while she is
replenishing reserves post-egg-laying and provisioning herself for the second incubation shift. The second shift, during which the male seeks to replenish his reserves
is likewise affected by krill availability but this time is several weeks later in the
year. On a longer time-scale we have breeding success (CEMP Index A6), monitored throughout the season, and fledging weight (CEMP Index A7). Together
these indicate how successful the adults have been in finding krill and bringing it to
the colony to provision the chick over the same period.
Using the icefish analogue when considering a still longer time-scale, the reproductive status of the fish is integrating krill availability over much of the year,
212
Krill: Biology, Ecology and Fisheries
although the precise form of that relationship is still not fully described. In addition
fish survey data can provide age structured population information with which to
describe demographic changes in the icefish population. The analogous CEMP
indices for penguins are population size and demography (CEMP Indices A3 and
A4 respectively).
Using these indices it should be possible to construct a series of models which
integrate over progressively longer time periods. The starting point would be the
short and medium term CEMP indices. Using the mass at first arrival, incubation
shift, diet and foraging trip indices a picture could be developed to infer local krill
status from May or June through to the end of the breeding season. The conceptual
framework for such a model is shown in Fig. 8.6. This model could then be incorporated into an age-structured model in order to determine population trajectories.
Such an approach has obvious advantages over the simple amalgamation of indices
into CSI because it is taking into account the ecological changes in the system.
Thus far in this chapter the consideration has been largely at the level of
predator±prey interactions. However one of the main reasons for undertaking this
type of study is to understand how commercial fishing might indirectly impact the
predators by affecting the abundance of their prey.
Fig. 8.6 Conceptual framework indicating the integration period for CEMP penguin indices. The
foraging trip duration and chick diet are both multiple observations during the course of the season.
The other indices are from single observations but at different times during the season.
Ecosystem Dynamics Involving Krill
8.6
213
Predator fisheries overlap
The introduction of reporting of Antarctic krill catches by fine-scale rectangles, half
a degree of longitude by one degree of latitude, demonstrated that the bulk of the
catch was being taken on the shelf or at the shelf break (Everson & Goss, 1992).
These localities were the same as those where penguins forage during the chick
rearing period. This then raised questions regarding the degree of overlap and
impact between fishing and dependent species foraging.
No simple procedure has been proposed to describe adequately this interaction
so, in order to understand the process we need to break it down into manageable
components. For simplicity I consider the problem from two perspectives, qualitatively in terms of time and space, and quantitatively in terms of the amounts of
krill taken by the fishery in relation to the requirements of dependent species.
Penguins during the breeding season provide a good example of a `central place
forager' since they are constrained by the range to which they can venture in search
of food and return in good time to feed a chick. Thus for any given colony the
foraging area can be described by the segment of a circle whose radius is the
maximum foraging range. This has been used to define a critical period distance
(CPD) for dependent species during the breeding season. Foraging activity can be
determined by direct observations from research vessels at sea leading to the
identification of particular foraging localities as shown in Fig. 8.7. It is also possible,
due to the miniaturisation of instrument packages, to monitor individual animals at
sea. This has been done for fur seal and to a lesser extent, due to the smaller size of
the animals, for penguins. Outside of the breeding season, it is assumed that the
birds forage over a much greater area; this may not be true necessarily but since they
are then not constrained by the need to return to a breeding site, they are able to
search for food in other localities.
The situation for fish is somewhat different. The mackerel icefish are known to be
restricted to the continental shelf region. It is not known to what extent feeding
migrations take place so spatially it has to be assumed that the whole shelf region is a
reasonable descriptor of their feeding range at all times of year. In Canadian waters,
hake are found on the shelf and slope regions (Simard & Mackas, 1989). Off the
coast of Japan there are pollock feeding on krill that are carried onto the shelf
(Endo, Chapter 7).
Several approaches to the problem of determining the relationship between
predator foraging and commercial fishing have been considered by CCAMLR.
These were defined in terms of the type of overlap.
. Precautionary overlap, which in considering the largest spatial scale is intended
to cover the whole distribution of krill and all krill predators. It is covered by the
krill yield model described earlier in this chapter and by Miller & Agnew, in
Chapter 12).
. Potential overlap, is on a very broad scale such that local overlaps or separations
Outer limit of survey
So
uth
(a)
Ge
org
ia
Outer limit of survey
Ecosystem Dynamics Involving Krill
215
between predators and the fishery may be missed or misrepresented. This has
been considered by CCAMLR as the critical period distance (CPD). The index is
currently calculated as the krill catch within 100 km of predator colonies during
the period December to March. It is not a measure of competition between
predators and the fishery, but is a simple expression of potential niche overlap.
Ichii et al. (1994a, b) considered the spatial distribution of colonies and catches in
the Antarctic Peninsula region. They showed that although a high proportion of
the catch may be taken within the foraging distance of land-based colonies, those
colonies containing the largest numbers of penguins were not adjacent to the
main fishing grounds. In order to make further progress with this type of work a
more refined approach is needed.
. Realised overlap, in which fine-scale overlap is measured but without taking
account of any movement of krill through the region. To describe this a simple
standardised index, such as Schroeder's index, has been used (SC-CAMLR,
1997). This has the form:
It ˆ 1 ÿ 0:5
X
j pi;t ÿ qi;t j
Where pi,t is the proportion of krill consumed by a predator in grid square I
during time period t and qi,t is the proportion of krill taken by the fishery in grid
square I during time period t. The index ranges from It = 0, indicating no spatial
overlap during period t to It = 1, complete overlap. Currently this index is used by
CCAMLR where the grid square is defined as the fine-scale rectangles already
mentioned above.
This approach has considerable merit in defining the localities and times during
which the interaction is likely to be most significant. However, since the two proportions are determined independently and without reference to the other component, care needs to be exercised in interpreting the results. Thus, to take an
extreme example, if the fishery were taking less than 1% by mass of the krill
required by penguins and all those krill were caught from a high index foraging
locality, the index would have a high value and yet no problem might exist. Some
mechanism is therefore needed to incorporate the total available krill into the index.
Another approach that attempts to incorporate fishing impact, and in fact predates the use of the Schroeder's Index by CCAMLR, was described by Agnew &
Phegan (1995). For this index the whole area was divided into small units of 10 6 10
nautical miles and the proportion P of the total number of penguins that were likely
Fig. 8.7 (Opposite.) (a) Foraging distribution of Antarctic fur seals in the vicinity of South
Georgia determined from direct visual observations at sea (Madureira, 1992). (b) Foraging footprint for the macaroni penguin at South Georgia. The figure links at-sea observations made during
the same study as that for Fig. 8.7(a) but integrates that information with data on breeding colony
location and size. (Part (b) with permission from Trathan et al., 1999.)
216
Krill: Biology, Ecology and Fisheries
to feed in each unit estimated. This is multiplied by the total krill requirement of the
penguins to give a penguin krill requirement for the location Kp. The other component is the krill catch Kc over the same period in the same area. The index is the
product (Kp . Kc ) and increases where either predation pressure or fishing increases.
Even so this still does not get around the problem of how much krill is available
within the small units.
Dynamic overlap, whereby the interaction would be described by the functional
link between predators and the fishery, would take account of fine-scale vertical and
horizontal distributions of predators and the fishery and the availability of prey to
both resource users. Undoubtedly this is the most difficult and complicated scenario
but does provide the greatest insight into the functioning of the system. A model
that makes comparisons of penguin reproductive success and adult survival in the
absence or presence of a krill fishery was developed by Mangel & Switzer (1998).
Over the range of catch used in the study, the reductions in reproductive success are
essentially linear functions of krill catch. The reductions in offspring and parent
survival as determined by the model were mainly determined by how long the
fishing season lasts and the capacity for harvest, rather than when, during the season, the fishing commences. In their paper the authors note further refinements
which might be incorporated into the model by refining the prey field, including a
component for chick mortality related to krill biomass and finally allowing for
plasticity in the fledging period. Clearly other approaches are also possible which
take account of other indices of the dependent species as well as incorporating
components which describe more closely the krill environment.
In this chapter so far I have been considering interactions in time and space from
the perspective of a system that was more or less enclosed and which might change
from year to year and between summer and winter. Such a view may be appropriate
for an enclosed sea or lake but misses an essential variable in the ecosystems in
which krill are key components. This additional variable is the movement of krill
through a region and is often referred to as krill flux.
8.7
Krill flux
In Chapter 7 several instances are noted whereby krill are known to be carried into,
through, and out of key regions. The amount of krill taken out of a region by
dependent species during the breeding season can be viewed as one estimator of
krill production in that region. Viewed from the perspective of krill production, that
amount of krill arose as a result of the balance of influx to and efflux from the region
allied to production by those krill present within the region. Consideration of the
flux of krill is therefore a further complication that must be included in the consideration of models from which management advice is developed. Krill are able to
move actively about in the water column; Watkins (Chapter 4) has highlighted their
swarming and vertical migratory capabilities. In spite of this their ability to move
Ecosystem Dynamics Involving Krill
217
over significant distances against major ocean currents is probably not great. Taking
a simple approach to addressing this problem, which might appear as something of a
contradiction in terms, we might assume that krill are passive particles that can
migrate vertically.
Everson & Murphy (1987) followed krill patches along the north side of the
Bransfield Strait and noted that they were moving at about the same rate as the
current in that region. The movements of the commercial fleets off the north coast
of the South Shetland Islands indicate that the fishing vessels are following patches
along the coast (Endo & Ichii, 1989). These movements appear to be at about the
same rate as the water movement in that region.
As an extension of this type of study, Murphy (1995) developed a model of the
South Georgia system in which the local prey abundance was regulated by a continuous, hydrodynamically mediated, supply rate and the concentrating effects of
abiotic±biotic interactions. Using estimates of annual predator demand and krill
concentration in the area the relationship between flow rate and depletion in prey
concentration as a function of range from the colony was estimated. The model
indicated that concentrating factors need to be large to produce the build up of krill
densities of the order estimated to occur in the South Georgia area. As might be
expected those predators which were restricted to the inshore region encountered
the greatest changes in prey abundance. This was further exacerbated when random
fluctuations in the inter-annual prey availability were introduced.
This approach was extended further by Murphy et al. (1998) by an examination of
the effects of inter-annual variation in year class strength on the standing stock and
also teleconnections between major areas. Taking information from the Fine
Resolution Antarctic Model (FRAM) to define the deep ocean current field, particles were seeded into the circumpolar current and their drift modelled across the
Scotia Sea. The rates of movement, whereby particles would take 140 to 160 days to
travel from the Antarctic Peninsula to South Georgia were in line with the timing of
arrival of different life history stages of krill across the region (Hofmann et al.,
1998). These results are consistent with the results from drogued buoys that took 110
to 120 days to drift over the same area.
Arising from this we can see that krill flux is operating in two major ways. On the
large scale, in deep oceanic water, the flux appear's to be consistent with estimates
of flow rate. Locally on the shelf the picture is much more complicated with krill
being concentrated in eddy systems and remaining there for extended periods, in
some instances of the order of months.
Much of this section has been concerned with highlighting differences due to
movement through and between regions. In spite of this large-scale movement there
is an underlying consistency over the locations where krill concentrations are found.
This can be seen in the distribution of whale catches during the 1920s (Everson,
1984). More recently it has been possible to identify what are becoming traditional
fishing grounds. For example at South Georgia the fleets regularly begin fishing at
the shelf break to the north-east of Cumberland East Bay (Murphy et al., 1997;
218
Krill: Biology, Ecology and Fisheries
Trathan et al., 1998), while in the South Shetlands the bulk of the fishing is along the
north side of King George Island up to Elephant Island (Ichii & Naganobu, 1996).
This level of consistency allied to the distribution of dependent species highlights
the need for an ecosystem approach to the management of krill, the starting point of
this chapter.
8.9
Selectivity
The preceding discussion has taken no account of the size or age of the krill being
taken by dependent species. For those krill species whose lifespan is less than a year
this is probably of little consequence for assessments. However, in situations where
several age or size classes are present this presents a further dimension for consideration.
Recent developments in sampling and analysis have meant that it is possible to
obtain information on the size distribution of the krill eaten, either from direct
measurements from stomach contents or scat samples (Reid et al., 1999). These
provide a good indication of the size, sex and in some cases maturity stage of the
krill eaten. This is important information to analyse in developing energy budgets
for the dependent species. The samples provide good indications of what the
predators actually consumed although they are not necessarily representative of the
size composition of the krill that were present or available within the foraging range.
The question then arises whether the predators are concentrating on taking a particular type of krill. Such a question does not have a simple answer because it will be
dependent on the type and amount of krill that are available within range of the
predator. Thus when krill are scarce the predator is likely to `take what it can find'
whereas when krill are plentiful the predator has some degree of choice.
It is possible to obtain samples from land-based predators with reasonable ease
and thus build up a short time-scale temporal picture of change through a season.
This needs to be checked against some direct and independent estimates of the krill
size distribution at intervals to determine the extent of selectivity. A study of this
type was undertaken by Reid et al. (1999) who was able to follow size classes
through a summer season (Fig. 8.8). As well as showing the growth of the dominant
size class through the season, part of the study compared the size distribution of krill
taken in net hauls from a short part of the season. This study shows two important
points. The first is that a single indication of the length±frequency distribution of the
krill being taken can be a very poor indication of the situation at other times during
the breeding season. The second point raises the question of whether the fur seals
were feeding on krill that were essentially passing the island as if on a conveyer, or
else whether the predation effect was influencing the size distribution of the krill
that were concentrated for a period in the region. This last point is important
because it raises the question of the direct influences of predation on krill mortality
rates.
Ecosystem Dynamics Involving Krill
219
Antarctic Fur Seal
28-Dec-93
03-Jan-94
10-Jan-94
Scientific Nets
18-Jan-94
Gentoo Penguin
24-Jan-94
30
38
46
54
Macaroni Penguin
62
Krill length (mm)
31-Jan-94
06-Feb-94
30
38
46
54
62
Krill length (mm)
13-Feb-94
20-Feb-94
30
38
46
54
62
Krill length (mm)
27-Feb-94
06-Mar-94
30
38
46
54
62
Krill length (mm)
Fig. 8.8 Sequence of krill length frequency distributions from Antarctic fur seals, scientific nets,
gentoo penguins and macaroni penguins during the period from December 1993 to March 1994.
(With permission from Reid et al., 1999.)
Providing the information on krill length distribution can be linked to predator
numbers and feeding intensity it can be compared to the amounts of krill present
from direct estimates. This then opens the way to developing more sophisticated age
structured models of the type outlined by Everson (1977).
8.10
Ecosystem monitoring and management
The concept of an ecosystem approach to the management of resources is on the one
hand intuitively obvious and on the other hand, due to the complexity of the system,
220
Krill: Biology, Ecology and Fisheries
extremely difficult to define. Even though the concept has been around for many
years ± the BIOMASS programme (El-Sayed, 1977) from which CCAMLR
developed was the earliest formal fisheries approach ± there is still some way to go
before it can be considered to be properly implemented. Within CCAMLR the
Working Group on Ecosystem Monitoring and Management (WG-EMM) (SCCAMLR, 1995) has been developing the concept of an ecosystem assessment. In
this context an ecosystem assessment would consist of two parts:
(1)
(2)
an analysis of the status of key biotic components of the ecosystem, and
a prediction of the likely consequences of alternative management actions for
the future status of these components.
In terms of ecosystem assessment, WG-EMM in 1995 agreed that:
`status would need to include not only the points necessary for a single species
assessment, which are:
current abundance and productivity of the harvested species, with abundance
related to some level prior to the onset of exploitation; and
if possible, the relationships (links) between these quantities and the state of
the environment.
but also points related to dependent species which may be summarised as:
current abundance of dependent species (usually expressed as breeding
population size or an index thereof) in relation to previous values, where
possible in conjunction with data on current and recent adult survival and
recruitment rates.'
These components can be expressed slightly differently as the following series of
questions which address problems of krill availability and dependent species
demography:
(1)
(2)
(3)
(4)
Is the availability of krill changing?
Are populations of dependent species in decline?
How much krill is required by the dependent species?
What is the extent of overlap between krill fishing and the foraging by
dependent species?
Addressing each of these questions in turn should provide a structure within
which a management regime can operate. Again I take as an example the
Southern Ocean and in particular CEMP. The CEMP parameters can be broadly
divided into two categories, those that indicate the availability of krill to the
predator and those that help interpret changes in predator demography. Such a
distinction, however, should not be considered as absolute, it is convenient in providing a breakdown of the information on the dependent species so that it can be
Ecosystem Dynamics Involving Krill
221
interpreted in a manner which will lead to the development of conservation
measures.
Moving through the breeding season of Macaroni penguin, one of the CEMP
species, we can establish a chronology whereby krill availability can be inferred. At
this stage, as indicated in Fig. 8.6, the consideration is purely temporal.
At the start of the breeding season the penguins arrive ashore. The total mass of
each individual on arrival will indicate how well they have been able to feed in the
weeks prior to arrival at the breeding site. Natural mortality will have had some
effect on the population but providing the recruitment balances deaths the numbers
arriving will be stable. Although mortality and recruitment are likely to be affected
most strongly by predation due to higher trophic levels, these may be mediated by
feeding conditions experienced by the penguins. The arrival mass of individuals and
the number of birds arriving at the site can therefore be taken as providing a longterm and medium-term index of krill availability over the winter months; the
temporal scale at this stage is coarse, of the order of months. The establishment of
the nest, egg laying and departure date of the female to feed, are all dependent upon
how the bird has fared in finding food over the recent past prior to its arrival back at
the colony. The temporal scale is shorter here when using the parameters to provide
an index of krill availability. The return of the female to end the first incubation shift
is critical, and her condition and the date at which she returns will be dependent on
how well she has been able to provision over the period. The temporal scale of the
interactions becomes progressively shorter as we move through the remainder of
the incubation phase and into chick rearing. By this time the temporal scale is down
to days. Difficulties in obtaining krill over this period will result in chick mortality
but if the situation improves then those chicks that survive are likely to achieve a
normal fledging mass.
This overview demonstrates that from the dependent species perspective the
temporal scale over which krill availability is assessed needs to be varied. Accepting
the need for this flexibility on the temporal scale necessitates that, because the
dependent species has a finite maximum swimming speed, the spatial scale is also
varying. However, providing we can accept that the dependent species is free to
forage at will within the temporal framework imposed by its breeding cycle this does
not matter because it is the temporal scale that is of primary importance. The spatial
scale is important in determining a `foraging footprint' and is discussed later.
Using this process we can develop rolling series of composite indices, the composition and temporal scale of which changes through the season. Examination of
the patterns of these indices for different years should provide indications of patterns within season and also, when there were serious breeding failures, the pattern
which led up to the failure. It is important to note that the time-scale during which
these indices are combined together will be species dependent, a feature which
although adding complication does at the same time provide an extra dimension by
allowing inter-species comparisons to be made. An analysis of these rolling indices
for the different species over a series of years will highlight those parts of the season
222
Krill: Biology, Ecology and Fisheries
when krill availability was affecting the dependent species. Combining this information with that obtained from studies on foraging range, analysed in a similar
manner to that shown in Fig. 8.6, provides a picture of the krill availability in time
and space through the breeding season. This information can be used to extend
knowledge on the distribution of krill obtained by directed surveys in the vicinity of
the breeding site.
The combined indices while providing a good indication of the amount of krill
available to the predators do not in this form provide indications of demographic
changes. This leads into the second question `Are populations of dependent species
in decline?'; what is needed here is the traditional population demographic information whose key components are population size, growth and mortality. These can
be determined from regular estimates of population size and a consideration of how
the CEMP-type parameters might influence the values. It is, however, important to
define the `population' in terms of practical management units. Thus although there
may be significant mixing between breeding colonies of dependent species at, for
example, adjacent island groups it may be desirable to separate them due to the
differing availability of krill experienced at the two localities. An example might be
the islands of the South Shetlands. Such a division might be determined by the
extent of the foraging footprint of key predators.
Regular and routine monitoring of population size is clearly indicated as the best
approach to this question. However this requirement needs to be set against the cost
in undertaking such a task and although it would be ideal to undertake such a task
annually that approach may not be cost effective. Reducing the frequency to a
three- or five-year interval means that there will be a greater period of time before a
downward trend can be positively identified. That is clearly a management risk that
falls outside of this chapter. Direct estimation of population size is not the only
approach because the same CEMP parameters can be used to investigate variation
in, for example, survivorship of adults, survivorship of chicks through to first
breeding. Essentially this is taking the basic components traditionally used in fish
stock assessment, recruitment, mortality and production and providing estimators
derived from direct observations.
Answers to the second question when considered in conjunction with the energetic requirements and dietary information lead to estimation of the amount of krill
required by the dependent species, the third question. Croxall et al. (1984) provided
estimates of food consumption at regular time intervals throughout the year.
Refinement of such estimates and examining them in conjunction with foraging
activity provides the `average' krill consumption term in the Agnew & Phegan
(1995) model previously mentioned. Studies on foraging behaviour have also been
developed to indicate the area over which the predators move in search of food, the
`foraging footprint' for the dependent species (Trathan et al., 1999). This foraging
footprint (Fig. 8.7b) will vary during the year due to the constraints imposed by the
breeding cycle. This spatio±temporal component leads naturally into the fourth
question, the extent of overlap between krill fishing and dependent species foraging.
Ecosystem Dynamics Involving Krill
223
The Schroeder index and the Agnew & Phegan model provide indices of overlap
and predation pressure on the krill. There is a third component which would assist in
determining whether fishing is likely to be having an effect on krill locally and that
is a simple ratio of the amount of krill taken by commercial fishing compared to
the amount required by dependent species. When this fishing to predation index,
(Kc /Kp) increases it indicates that fishing is taking a larger proportion of the
available krill and consequently fishing is more likely to have an impact on the
dependent species. It is important to note that this index does not necessarily
indicate that any management action is required but rather provides a warning that
an effect might be noticeable in some of the predator demographic parameters.
Two further points need to be made on this theme. Firstly that because the
krill sampling techniques differ widely it is important that the estimates of krill
standing stock and production are properly estimated. Commercial fishing estimates krill by applying a conversion factor to the amount of product, research
surveys provide estimates based on scaling acoustic signals while figures for
dependent species are derived from a combination of population size, energy
requirement and diet; all of these are subject to different degrees of imprecision.
The second point relates to prey switching by predators. In times of krill scarcity
predators are likely to take alternative prey. A good example is the Antarctic fur
seal at South Georgia, which feeds on fish during periods of krill scarcity. This is
thought to have a major impact on the mackerel icefish population (Everson et
al., 1999) but since this goes beyond the direct effects of krill in the ecosystem it
is not considered further here.
With the practical framework to providing an ecosystem assessment outlined
above we are now in a position to consider how this can be taken forward as
management advice.
8.11
An ecosystem approach to krill management
The problem of krill management in an ecosystem context can be reduced to two
key questions. The over-riding and first question is: `Taking account of the
requirements of dependent species, is the krill fishery sustainable in the long term?'
This question is adequately addressed by the KYM and providing there is a maximum limit for the krill fishery in place no further comment is needed.
The second question is: `Are dependent species being adversely affected by
krill fishing?'. This in turn breaks down into a series of questions and action
points as shown in Fig. 8.9. The primary consideration is to determine whether
there is any evidence for a decline in the populations of dependent species. This
may come from the direct estimation of population size but might also appear as
an increase in mortality rate before a significant trend was recognisable in population size.
A decline in population size does not itself indicate that the cause was com-
224
Krill: Biology, Ecology and Fisheries
Fig. 8.9 Decision processes incorporating information from dependent species into a mechanism
to provide advice for management of a krill fishery.
mercial fishing so the next level of enquiry should be towards investigating the
various possibilities. Analysis of indices which relate to krill availability will indicate whether the cause lies with the ability of the predators to find krill. Such an
investigation needs to take account of the oceanography of the region to determine whether underlying cause might be due to variation in oceanic circulation
or production. The investigation should also take account of the degree of overlap between fishing and predation as indicated by the Schroeder, Agnew &
Phegan and fishing±predation indices. With this information it is then possible to
provide advice on locations and times of year when fishing pressure might need
to be reduced. Such information would translate directly to conservation
measures for the krill fishery in question. There is also the possibility that the
underlying cause for the decline in the population of a dependent species might
not be due to fishing but environmental change, either natural or man-induced,
such as through global warming. Considerations of that type fall outside the remit
for this chapter but are nonetheless vitally important topics in a global management context.
Using the decision processes envisaged in Fig. 8.9 we have a structure within
which a science plan can be formulated that can provide advice for management of
krill fisheries in an ecosystem context. Further refinements are clearly possible but
establishing this first link would be a major step in fisheries management.
Ecosystem Dynamics Involving Krill
225
References
Agnew, D.J. & Phegan, G. (1995) A fine scale model of the overlap between penguin foraging
demands and the krill fishery in the South Shetland Islands and Antarctic Peninsula.
CCAMLR Sci. 2, 99±110.
Beddington, J.R. & Cooke, J.G. (1983) The potential yield of fish stocks. FAO Fisheries
Technical Paper 242 FIRM/T242, 53 pp.
Boyd, I.L. & Croxall, J.P. (1992) Diving behaviour of lactating Antarctic fur seals. Can. J.
Zoology 70, 919±28.
Boyd, I.L., McCafferty, D.J. & Walker, T.R. (1997) Variation in foraging effort by lactating
Antarctic fur seals: response to simulated increased foraging costs. Behav. Ecol. Sociobiology 40, 135±44.
Bradstreet, M.S.W. & Brown, R.G.B. (1986) Feeding ecology of the Atlantic Alcidae. In The
Atlantic Alcidae (D.N. Nettleship & T.R. Birkhead, eds). Academic Press, London.
Butterworth, D.S. & Thomson, R.B. (1995) Possible effects of different levels of krill fishing
on predators ± some initial modelling attempts. CCAMLR Sci. 2, 79±98.
Butterworth, D.S., Punt, A.E. & Basson, M. (1992) A simple approach for calculating the
potential yield of krill from biomass survey results. SC-CAMLR Selected Scientific Papers
8, 207±217.
Butterworth, D.S., Gluckman, G.R., Thomson, R.B., Chalis, S., Hiramatsu, K. & Agnew, D.J.
(1994) Further computations of the consequences of setting the annual krill catch limit to a
fixed fraction of the estimate of krill biomass from a survey. CCAMLR Sci. 1, 81±106.
Constable, A.J. & de la Mare, W.K. (1996) A generalised model for evaluating yield and the
long-term status of fish stocks under conditions of uncertainty. CCAMLR Sci. 3, 31±54.
Croxall, J.P. & Lishman, G. (1987) The food and feeding ecology of penguins. In Seabirds.
Feeding Ecology and Role in Marine Ecosystems (J.P. Croxall, ed.), Cambridge University
Press, Cambridge.
Croxall J.P., Prince, P.A. & Ricketts, C.R. (1984) Impact of seabirds on marine resources
especially krill, of South Georgia waters. In Seabird Energetics (G. Causey Whittow & H.
Rahm, eds), pp. 285±317. Plenum, New York.
El-Sayed, S.Z. (ed.) (1977). Biological Investigations of Marine Antarctic Systems and Stocks,
Vol. 1, 79 pp. Scott Polar Research Institute, Cambridge, England.
Endo, Y. & Ichii, T. (1989) CPUEs body length and greenness of Antarctic krill during 1987/
88 season in the fishing ground north of Livingston Island. Scientific Committee for the
Conservation of Antarctic Marine Living Resources, Selected Scientific Papers. SCCAMLR-SSP/6, pp. 323±45.
Everson, I. (1977) The Living Resources of the Southern Ocean. Food and Agriculture
Organisation of the United Nations, Southern Ocean Fisheries Survey Programme. GLO/
SO/77/1, 156 pp.
Everson, I. (1983) Estimation of Krill abundance. In On the biology of Krill Euphausia
superba. In Proceedings of the Seminar and Report of the Krill Ecology Group, Bremerhaven, 12±16 May 1983. (S.B. Schnack, ed.), Berichte zur Polarforschung, Sonderheft 4,
December 1983, pp. 156±68.
Everson, I. (1984) Marine interactions. In Antarctic Ecology, Vol. 2., Chapter 14 (R.M. Laws,
ed.), pp. 785±819). Academic Press, London.
Everson, I. Klock, K.-H. & Ellison, J. (2000) Inter-annual variation in the gonad cycle of the
mackerel icefish. J. Fish Biol.
226
Krill: Biology, Ecology and Fisheries
Everson, I. & de la Mare, W.K. (1996) Some thoughts on precautionary measures for the krill
fishery. CCAMLR Sci. 3, 1±11.
Everson, I. & Goss, C. (1992) Krill fishing activity in the southwest Atlantic. Antarctic Sci. 3,
351±58.
Everson, I & Murphy, E.J. (1987) Mesoscale variability in the distribution of krill Euphausia
superba. Mar. Ecol. Prog. Ser. 40, 53±60.
Everson, I., Parkes, G., Kock, K.-H. & Boyd, I.L. (1999) Variation in standing stock of the
mackerel icefish Champscephalus gunnari at South Georgia. J. Appl. Ecol. 36, 591±603.
Gulland, J.A. (1971) The Fish Resources of the Ocean. Fishing News (Books), West Byfleet,
Surrey, for FAO, 255 pp.
Hofmann, E.E., Klinck, J.M., Locarnini, R.A., Fach, B., & Murphy E. (1998) Krill transport in
the Scotia Sea and environs. Antarctic Sci. 10, 406±415.
Ichii,T. & Naganobu, M. (1996) Surface water circulation in krill fishing areas near the South
Shetland Islands. CCAMLR Sci 3, 125±36.
Ichii, T., Naganobu, M. & Ogishima, T. (1994a) An assessment of the impact of the krill
fishery on penguins in the South Shetland Islands. CCAMLR Sci 1, 107±113.
Ichii, T., Naganobu, M. & Ogishima, T. (1994b) A revised assessment of the impact of krill
fishery on penguins in the south Shetland Islands. Document WG-Joint-94/17, 20 pp.
CCAMLR, Hobart.
Kock, K.-H. (1992) Antarctic Fish and Fisheries. Cambridge University Press, Cambridge.
Kock, K.-H. , Wilhelms, S., Everson, I. & GroÈger, J. (1994) Variations in diet composition and
feeding intensity of mackerel icefish Champsocephalus gunnari at South Georgia (Antarctic). Mar. Ecol. Prog. Ser. 108, 43±57.
Laws, R.M. (1984). Seals. In Antarctic Ecology (R. Laws, ed.), pp. 531±616. Cambridge
University Press, Cambridge.
Love, R.M. (1988) The Food Fishes their intrinsic variation and practical implications. Van
Nostrand Reinhold, New York.
Madureira, L. St.P. (1992) Application of a dual frequency echosounder to the identification
and quantification of Antarctic krill Euphausia superba distribution around South Georgia
in relation to its principal predators. PhD dissertation, University of Cambridge, July 1992.
167 pp.
Mangel, M. & Switzer, P.V. (1998) A model at the level of the foraging trip for the
indirect effects of krill (Euphausia superba) fisheries on krill predators. Ecol. Model.
105, 235±56.
Murphy, E.J. (1995) Spatial structure of the Southern Ocean ecosystem: predator-prey linkages in Southern Ocean food webs. J. Anim. Ecol. 64, 333±47.
Murphy, E.J., Trathan, P.N., Everson, I., Parkes, G. & Daunt, F. (1997). Krill fishing in the
Scotia Sea in relation to bathymetry, including the detailed distribution around South
Georgia. CCAMLR Sci. 4, 1±18.
Murphy, E.J., Watkins, J.L., Reid, K. et al. (1998) Interannual variability of the South Georgia
marine ecosystem: biological and physical sources of variation in the abundance of krill.
Fish. Oceanogr. 7, 381±90.
Reid, K., Watkins, J.L., Croxall, J.P. & Murphy, E.J. (1999) Krill population dynamics at
South Georgia 1991±1997, based on data from predators and nets. Mar. Ecol. Prog. Ser.
177, 103±114.
SC-CAMLR (1992) Report of the Eleventh meeting of the Scientific Committee for the Con-
Ecosystem Dynamics Involving Krill
227
servation of Antarctic Marine Living Resources, pp. 426±27. SC-CAMLR, Hobart,
Australia.
SC-CAMLR (1995) Report of the Fifteenth meeting of the Scientific Committee for the Conservation of Antarctic Marine Living Resources, SC-CAMLR, Hobart, Australia.
SC-CAMLR (1997) Report of the Sixteenth meeting of the Scientific Committee for the
Conservation of Antarctic Marine Living Resources. SC-CAMLR, Hobart, Australia.
Simard, Y. & Mackas, D.L. (1989) Mesoscale aggregations of Euphausiid sound scattering
layers on the continental shelf of Vancouver Island. Can. J. Fish. Aquat. Sci. 46, 1238±49.
Trathan, P.N., Everson, I., Murphy, E.J. & Parkes, G. (1998) Analysis of haul data from the
South Georgia krill fishery. CCAMLR Sci. 5, 9±30.
Trathan, P.N., Murphy, E.J., Croxall, J.P. & Everson, I. (1999) Use of at-sea distribution data
to derive potential foraging ranges of macaroni penguins during the breeding season. Mar.
Ecol. Prog. Ser. 169, 263±75.
Chapter 9
Krill Harvesting
Taro Ichii
9.1
Introduction
A krill fishery was first conducted for Meganyctiphanes norvegica in the Mediterranean in the 19th century where the catch was used as bait (Fisher et al., 1953).
In recent years, large-scale krill fishing has targeted Euphausia superba in the
Antarctic and E. pacifica in coastal waters off northeastern Japan, with smaller scale
harvesting of E. pacifica off western Canada (Eddie, 1977; Everson, 1978; Nicol &
Endo, 1997). Minor, intermittent fisheries for Thysanoessa inermis and for E. nana
off Japan, were reviewed by Nicol & Endo (1997). This chapter presents information on krill harvesting in the former three fisheries, including the history of each
fishery, fishing strategy, fleet organisation, characteristics of fishing grounds,
detection and catching of krill swarms and the primary concerns of fishermen.
Finally, comparisons of the three krill fisheries highlight the characteristics of each.
9.2
Antarctic waters (Euphausia superba)
History of the fishery
Ever since a high abundance of E. superba became apparent (Marr, 1962; Miller &
Agnew, Chapter 12) speculation existed that the species might be harvested commercially. During the 1960s, it was estimated that a total catch of upwards of 150
million tonnes of krill might be taken. This amount, often referred to as a `krill
surplus' created by the great reduction in baleen whale stocks (Laws, 1977) caused
heightened interest in an Antarctic krill fishery. However, the principle reason for
the initiation of harvesting was probably the displacement of fishing fleets from
Japan, the former USSR, and other Eastern Bloc countries from 200 n. miles
Exclusive Economic Zones, as a result of developments in the Law of the Sea
negotiations in the 1960s and early 1970s (Kock, 1992). In Japan, krill from the
stomachs of baleen whales (Plate 4, facing p. 182) caught by whaling were initially
used as food for aquaculture, a practice which also promoted the development of the
Antarctic krill fishery (Kaneda, personal communication).
The commercial krill fishery started in the early 1970s (Fig. 9.1). Pelagic fishing
228
Krill Harvesting
229
Fig. 9.1 Annual catches of Antarctic krill in the Atlantic (Statistical Area 88), Indian (Area 58)
and Pacific (Area 88) sectors of the Southern Ocean (CCAMLR, 1990a; 1990b; 1997; SC-CAMLR,
1997). Statistical Areas shown in Fig. 9.2.
activities from former Eastern Bloc countries were heavily subsidised which
facilitated fishing operations in waters where vessels from countries with market
economies, and operating without a significant subsidy, had great difficulties fishing
economically (Kock, 1994). The fishery reached a peak in the early 1980s when over
half a million tonnes were caught. However, catches dropped substantially thereafter due to problems in krill processing and increased interest in finfish (Kock,
1992). From 1986±87 to 1990±91, annual catches became stabilised at around
350 000 to 400 000 t, the fishery accounting for about 13% of the world catch of
crustaceans at that time (Nicol, 1991). Catches declined again in the 1991±92 season
and even more dramatically to approximately 87 000 t in the 1992±93 season, when
economic considerations caused the states of the former USSR to discontinue krill
fishing. Thus, most of the catch until 1991±92 was taken by the USSR, the dominant
nation thereafter being Japan (Table 9.1). Other nations, South Korea, Poland,
Chile, Bulgaria, Latvia, Panama, Spain, East Germany and UK, have also been
involved in the krill fishery, but have always been `minor players' compared to the
former USSR and Japan.
The krill fishery in the Southern Ocean is regulated under the auspices of the
Commission for the Conservation of Antarctic Marine Living Resources
(CCAMLR), which came into force in 1982 (Fig. 9.2) The annual fishing period for
the Antarctic is referred to as split-year, which extends from 1 July to 30 June.
However, in the South Georgia area (Subarea 48.3), as the fishing season falls in
±
±
±
±
94
46
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
Bulgaria
±
±
276
92
±
±
±
±
±
3752
1649
2598
3264
4063
5938
5394
4500
3679
6065
3261
3834
±
±
±
±
Chile
±
±
±
±
±
±
6
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
France
8
102
±
±
±
±
±
±
±
±
±
±
396
±
±
±
±
±
±
±
±
±
±
±
±
GDR
646
2676
4750
12 802
25 219
36 961
36 276
27 698
35 755
42 326
49 572
38 274
61 074
78 389
73 112
78 928
62 187
67 582
74 325
59 272
62 322
60 303
60 546
58 798
63 233
Japan
±
±
±
±
±
511
±
±
1429
1959
2657
±
±
1527
1525
1779
4039
1210
519
±
±
±
±
±
1623
Korea
Data from CCAMLR (1990a; 1990b; 1997) and SC-CAMLR (1997)
1973/74
1974/75
1975/76
1976/77
1977/78
1978/79
1979/80
1980/81
1981/82
1982/83
1983/84
1984/85
1985/86
1986/87
1987/88
1988/89
1989/90
1990/91
1991/92
1992/93
1993/94
1994/95
1995/96
1996/97
1997/98
Year
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
71
±
±
±
±
Latvia
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
141
495
±
±
Panama
21
6966
37
±
226
±
±
360
±
±
2065
1726
5215
7871
1275
9571
8607
15 909
7915
9384
20 610
19 156
15 312
±
±
Poland
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
151725
4249
965
±
±
±
±
Russia
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
2
±
±
±
±
S. Africa
Table 9.1 Reported catches of Antarctic krill by different nations fishing in the CCAMLR area 1973±1998.
±
±
±
±
±
±
±
±
±
±
±
±
±
450
±
±
±
±
±
±
±
±
±
±
±
Spain
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
61 719
6083
8852
48 884
20 056
4246
±
Ukraine
21 700
38 900
500
105 049
116 601
295 508
440 516
420 434
491 656
180 290
74 381
150 538
379 270
290 401
284 873
301 498
302 376
275 495
±
±
±
±
±
±
±
USSR
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
308
634
UK
Fig. 9.2 CCAMLR Statistical Areas, Subareas, and Divisions. With permission from the CCAMLR.
Krill Harvesting
231
232
Krill: Biology, Ecology and Fisheries
June and July, thus encompassing the end of one split-year and the beginning of the
next, the annual period is expressed as calendar years.
Fishing strategy
A number of difficulties have been associated with the E. superba fishery. Besides
the limited market for krill products, the greatest problem is distance; the fishery
being centred far from major ports and being undertaken in very inhospitable seas.
The geographical remoteness is reflected in the areas that have been fished for krill.
Although krill have a circumpolar distribution, as shown in Figs 9.3 and Table 3.3.2,
the fishery has tended to concentrate in areas close to other fisheries off South
America, South Africa and New Zealand. Thus, when krill fishing is less profitable,
the vessels engage in other fishing activities, such as for squid and finfish off South
America for the reminder of the year.
Krill quality is also a major concern. Important from the commercial view point,
E. superba products are graded by `greenness', body size and body colour. `Green'
krill are those which have been feeding intensively on phytoplankton, that accumulates in the frontal part, specifically in the hepatopancreas inside the carapace.
This is unimportant if peeled krill meat or meal is being produced, but `green' krill
are actively avoided if fresh-frozen or boiled-frozen products are required, owing to
the dirty appearance, unfavourable smell and inferior taste of products from the
former. `Green' krill tend to occur in the early austral summer (December±January)
(Kawaguchi et al., 1998).
In the Japanese fishery the krill are graded by the fishermen according to their
size: LL (larger than 45 mm), L (between 35 and 45 mm), and M (below 35 mm).
Generally, the LL size category, especially if it contains gravid females, is preferred
for both human consumption and sport fishing. It is also easier to peel larger krill.
However, because the smaller classes (L and M sizes) are in some demand for sport
fishing and aquaculture, it is advantageous to operate in fishing grounds where krill
show size segregation, allowing size-specific targeting to be employed (Kigami,
personal communication).
Krill are also graded according to body colour. Transparent krill, termed `white'
are firm with a pleasing appearance, whereas brick red/pink krill, termed `pink', are
flaccid and easy to crush. The former, more valuable than the latter, are more
frequently present from late in the austral summer (Ichii, 1987). Because of the
substantial variability in quality of E. superba, the prime fishing strategy is to harvest
the most marketable krill, rather than to concentrate solely on maximising the catch
(Kigami & Kaneda, personal communication).
Fishing fleet organisation
Principally two types of operational pattern occur, which for convenience can be
termed individual ship operation and closely cooperating fleet operation. The
Krill Harvesting
233
Japanese fishery has essentially been structured around the first type with stern
trawlers mostly of around 3000 to 4000 t GRT, although communications are
maintained between vessels belonging to the same company (Ichii, 1987). In contrast, the former USSR vessels, which were mostly 3800 t stern trawlers, used to
work in close cooperation, their fishing fleets usually operating in concert with
scouting vessels (Everson, 1978, 1988). The latter continually searched for new krill
concentrations and surveyed their properties (location, horizontal extension, catch
rate, krill size, `greenness') so as to inform the catching vessels (Dolzhenkov et al.,
1988)
Detailed information is available for the `mother-ship' operation conducted by
Japan in the early years as a subsidised venture from 1977±78 to 1981±82 (Shimadzu,
1984). It consisted of a mothership (8000 t class) with freezing and processing
capabilities, and 7 to 10 stern trawlers (349 t class) with boiling and freezing
facilities. One or two trawlers operated as `scouts' outside of the fleet's fishing
ground. Even though the trawlers were able to process krill, they frequently
transported catches to the mothership to make full use of the latter's more effective
processing capabilities. A small transportation boat, associated with the mothership,
received the cod-end from the trawler and then transferred it to the mothership, a
transporting procedure conducted so smoothly that little difference in quality
occurred between krill products made by the mothership and the trawlers. Economically, however, the mothership operation did not compare favourably with
individual trawler performances since the fleet typically caught 200±300 t/day in
total compared to the 50±70 t average for a 3000 t class vessel in those days, and
accordingly was discontinued (Shimadzu, 1984).
The remote operation of this fishery is supported by cargo vessels, which
frequently, twice a month during high-season, visit the trawlers to replenish food
and fuel supplies, and pick up krill products. When sea conditions are rough, cargo
transfer is carried out close to the ice edge or in the lee of islands so as to take
advantage of the calmer conditions there. These days, many of the trawlers are
based in foreign ports such as in Chile, Argentina and South Africa, so that both the
krill and other fisheries have become a much less complicated logistic operation.
Characteristics of the fishing grounds
Antarctic krill characteristically aggregate into concentrations, comprising many
swarms, and these are targeted by fishing trawlers (Butterworth & Miller, 1987;
Ichii, 1987). Since the concentrations are formed in limited areas, vessels make use
of a range of information to locate harvestable concentrations. Most important is
the historical distribution of good harvesting areas (Ichii, 1987). During searching
activity, surface water temperatures are monitored in order to detect oceanographic
fronts, which tend to hold krill aggregations for prolonged periods. Icebergs with
many resting seabirds or covered by bird droppings can also be indications of the
existence of nearby krill aggregations. Once such `indicators' have been found,
234
Krill: Biology, Ecology and Fisheries
vessels intensify their searching activities by increasing the frequency of directional
changes so as to locate the high density part of the krill concentration. Communication with other fishing vessels within the same fishing company enhances the
likelihood of finding high krill concentrations. The fishing vessels avoid operations
near the ice edge, because of the problems encountered when trawling in icy waters.
Historical fishing grounds thus exploited are divided geographically into three
regions: off Kemp Land, off Wilkes Land and the south-west Atlantic (Fig. 9.3). The
krill fisheries in these three regions have been combined with demersal fishing off
Africa, squid fishing off New Zealand and fishing off South America for squid and
finfish. The greatest krill catches have been taken between 508E±708E in Kemp
Land, 1108E±1508E in Wilkes Land, and 308W±658W, South Shetland Islands, South
Orkney Islands and South Georgia, in the south-west Atlantic. Some effort has been
extended over wider ranges than these limits indicate (Fig. 9.3). Good fishing
grounds are closely associated with ice-free continental and insular shelf break±
slope regions (Ichii, 1990; Everson & Goss, 1991). In the most recent years, almost
all vessels have operated in the south-west Atlantic.
Krill fishing off Kemp Land and Wilkes Land used to be conducted in the austral
summer, with fishing activity shifting from the offshore to the continental shelf
break as the ice edge retreated southwards (Fig. 9.4). In austral mid-summer, krill
trawling was concentrated along the continental shelf break. In the south-west
Atlantic, on the other hand, the krill fishery has recently operated from the austral
summer through to the winter (Fig. 9.5). The fishery begins in the South Shetland
Islands area in mid austral summer, moving north as the sea-ice advances through
the South Orkney Islands area to the northernmost fishing area around South
Georgia by austral mid-winter (Everson & Goss, 1991). The most popular fishing
grounds are formed near the shelf break of the respective islands.
The daily catch rate has tended to be highest, around 50±150 t/vessel, in the southwest Atlantic sector, an area that includes the South Shetland Islands, South Orkney
Islands and South Georgia. Lower catch rates, of around 30±90 t/vessel, were made
off Wilkes Land (Fig. 9.6), this suggests that krill are more abundant and the fishing
grounds more stable in the Atlantic sector than off Wilkes Land. The fishing
grounds are most likely influenced by complex interactions between krill behaviour
and environmental factors. Regarding the latter, the continental slope front is
considered to be responsible for the formation of fishing grounds off Wilkes Land
(Ichii, 1990). In the South Shetland Islands area, which has been intensively studied,
the coupling of oceanographic features with favourable feeding conditions is
important; the shear zones and sluggish currents caused by the insular shelf may
provide stable sites that retain krill. Since these sites also have abundant food
sources, krill may actively move to them, resulting in dense quasi-stable concentrations (Ichii et al., 1998).
Krill in the South Shetland Islands area show offshore±onshore segregation in
both size and maturity stages (e.g. Siegel, 1988; Siegel et al., 1997; Ichii et al., 1998).
Hence fishing vessels can select larger krill by operating in the offshore and slope
Krill Harvesting
235
Fig. 9.3 Distribution of Antarctic krill fishing grounds in relation to continental and insular
shelves and pack-ice edge during austral summer in the Antarctic. (a) Mean pack-ice edge in
February, ice-free continental shelf and Polar Front. (b) Fishing ground distribution.
236
Krill: Biology, Ecology and Fisheries
Fig. 9.4 Distribution of accumulated catches of Antarctic krill (tonnes/10±10 n. miles) from 1976/
77 to 1993/94 for each month off Wilkes Land. Maximum (Ð) and minimum (>) ± pack-ice edge in
each month during 1976/77±1993/94 are shown. Dotted lines indicate 500 m depth contours. Catch
data from Japanese fishery. Size of closed circles proportional to catch size, e.g. closed circles of
maximum diameter correspond to > 500 tonnes.
Fig. 9.5 Distribution of recent (1989/90±1996/97) accumulated catches of Antarctic krill (tonnes/
30±30 n. miles) for each month in the Southwest Atlantic. Maximum (± . ± . ±) and minimum (± ± ± ±) ±
pack-ice edge in each month during 1989/90±1996/97 are shown. Catch data from Japanese fishery.
Size of closed circles proportional to catch size, e.g. closed circles of maximum diameter correspond to
> 1000 tonnes. SS = South Shetland Islands, SO = South Orkney Islands, SG = South Georgia.
238
Krill: Biology, Ecology and Fisheries
Fig. 9.6 Interannual variations in average daily catch (tonnes/vessel) in each fishing area for
Antarctic krill. Data from Japanese fishery. SS: South Shetland Islands; SO: South Orkney Islands;
SG: South Georgia; WL: Wilkes Land; KL: Kemp Land. Calendar years in parentheses are for the
South Georgia area.
regions. Such selection is difficult in other areas since segregation is not apparent.
The size distribution of the krill caught during the austral winter in South Georgia is
distinctly different to that from other fishing grounds in that larger krill (>50 mm)
are usually absent. This may be because krill may shrink due to food shortages
during winter as Ikeda & Dixon (1982) suggested although it is more likely due to
the advection of small krill into the region.
Advantages in operating off Wilkes Land lay in the quality of krill caught, which
tended to be better than in the south-west Atlantic during the austral summer such
that there were fewer `green' krill, leading to a greater proportion of white firm
products rather than red±pink and soft ones (Ichii, 1987). In fact processing of the
fresh-frozen krill needed to be completed within three hours of capture, compared
with two hours in the Southwest Atlantic. These advantages did not compensate
however for the lower catch rates already noted above. By-catches of salps may also
be a problem in the Southwest Atlantic (Ichii, 1987).
Detection and catching of krill swarms
On the fishing grounds, the detection by echo-sounder and sonar leading to capture
of individual aggregations are the basic activities of the harvesting operation. Echosounders can detect swarms at most depths immediately below the vessels, but have
very limited horizontal range, no more than vessel width. Sonars, on the other hand,
Krill Harvesting
239
have an effective horizontal detection range of as much as 300 to 500 m when
swarms are dense and well below the surface, but only a limited vertical range that is
constrained by the transducer beam width, which could well miss swarms even
directly below the vessel.
Echo-sounders are the most important swarm detection device although in early
summer, when swarms generally are small, sonar is indispensable. In areas where
the number of swarms per unit area is so high that detection by sonar is unnecessary,
the main purpose of the latter is for determining the direction of the swarm
extension, so as to judge the optimal direction for towing (Ichii, 1987). In areas
where krill form dispersed aggregations greater than 3±4 km in chord length, their
densities are usually too low for sonar detection, location by echo-sounder being the
only method. Surface swarms, which occur regularly only in the half-light period
after sunset off Kemp and Wilkes Land during the early austral summer, can be
detected by the naked eye, but usually only by experienced observers (Ichii, 1987).
Fishing vessels usually concentrate on the larger and denser swarms for harvesting (Ichii, 1987). When a fishable swarm is detected, the vessel is manoeuvred
on to a course that will result in the swarm being intercepted by the trawl. When the
trawl is shot, final adjustments to the course are made by reference to sonar.
Adjustments to the trawl depth are made by tracking the swarm depth as seen on the
echo-sounder and net sounder. The latter, which relies on vertically scanning
transducers attached to the headline of the trawl net, has two principal functions: to
indicate net depth in relation to the surface and the quantity of krill that have
actually entered the net. For krill harvesting, a midwater trawl net with a mouth
opening of 500 to 700 m2 is used (Plate 5, facing p. 182). Since krill appear to take no
effective avoidance action to commercial nets, towing speeds are generally around
two knots. If wind speeds are more than 2±3 on the Beaufort scale (about 5±10
knots), the vessels trawl with the wind from behind.
Haul duration is generally kept to a time which will ensure a maximum catch of
between 7 and 10 t. This is for two reasons, firstly because with larger hauls, since the
krill are thin and easily crushed, product quality suffers, and secondly because
operations need to be linked to the vessel's processing capabilities. To maintain such
a catch level, fishermen use the net sounder to determine the optimal number of
swarms fished or else adjust the towing time (Fig. 9.7). When larger cohesive
aggregations of several hundred metres in horizontal length are trawled, a single
swarm is usually sufficient for the optimal catch. Even so, when widely dispersed
low-density aggregations extending over many kilometres are fished, it may take at
least an hour to reach the desired catch level.
Towing is conducted continuously throughout the day and night, enabling processing to be conducted in an unbroken sequence. Analyses of catch per unit of
effort (CPUE), where effort is determined by trawling time, and trawling depth
show different diurnal patterns between fishing grounds in the south-west Atlantic
(Fig. 9.8). In the South Shetland Islands area during the austral summer (January±
April), only slight diurnal changes in CPUE occurred, being larger in the daytime
240
Krill: Biology, Ecology and Fisheries
Fig. 9.7 Net recorder chart when a dense swarm was fished. Height of net was 37 m. 20 tonnes (too
much!) of Antarctic krill was caught in 14 min. of towing. Absolute density was estimated as 40 g/
m3. The fishing was conducted at 66803'S, 145853'E on 2 March, 1986.
(160±300 kg/min) than at night (100±240 kg/min). There was also a small and slight
diurnal change in trawl depth with fishing being deeper in the daytime (60±80 m
depth) than at night (40±50 m). At South Georgia during the austral winter months
of June through to August, on the other hand, a marked distinct diurnal change in
CPUE was apparent with much higher values in the daytime (220±700 kg/min) than
at night (80±380 kg/min). Trawling depths were distinctly deeper than those in the
South Shetland Islands area and showed substantial inter-annual variations, especially in the daytime where they varied from 50 to 250 m. It should be noted that
daytime trawling depths were significantly correlated with krill size (Fig. 9.9)
because the larger krill tend to descend deeper in daytime (also see Kawaguchi et al.,
1996). In the South Orkney Islands during the austral autumn months of April to
June, both CPUE and trawling depth show an intermediate pattern between those
in the South Shetland Islands and South Georgia. These differences in CPUE and
trawling depth among the three fishing areas may reflect both regional and seasonal
changes in distributional behaviour of krill. Thus, because of the generally higher
CPUE, winter krill fishing operations are not disadvantageous by comparison with
those in the austral summer, despite the rougher sea conditions encountered during
winter.
Primary concerns of the fishermen
On the south-west Atlantic fishing grounds fishermen are most concerned with the
quality of krill, since catching large quantities is still fairly straightforward. Lower
quality krill `green' and `red-coloured' can be avoided by operating during the late
austral summer, autumn and winter months. Recently, for example, Japan has
extended the fishing season into the austral winter season, so as to avoid catching
early season `green' krill and thereby increase the catch of colourless (white) krill,
which become available later in the season (SC-CAMLR, 1992; Kawaguchi et al.,
Fig. 9.8 Diurnal variations in average catch per towing time (a)±(c) and trawling depth (d)±(f) from 1989/90 to 1996/97 in each fishing area for Antarctic
krill in the Southwest Atlantic sector. SS: South Shetland Islands (January to April); SO: South Orkney Islands (April to June); SG: South Georgia (June
to August). Data from Japanese fishery.
Krill Harvesting
241
242
Krill: Biology, Ecology and Fisheries
Fig. 9.9 Relationship between average trawling depth at noon and average total body length of
Antarctic krill caught during the austral winter in the South Georgia area. Linear regression (solid
line) with 95% confidence limits (broken lines) are shown. y = 7521.4 + 16.5x, where y is trawling
depth (m) and x is body length (mm). (r = 0.94, p < 0.0005, n = 8.)
1998; Plate 2, facing p. 182). However, `body size', varying between years rather
than following a consistent seasonal pattern, is often a major problem. This is caused
by recruitment of E. superba showing great inter-annual variability (Siegel & Loeb,
1995), with subsequent growth to the largest size class (>50 mm) requiring a period
of 4±5 years (Siegel, 1987). Figure 9.10 shows substantial inter-annual variability in
the size composition of krill in catches for each fishing area, the patterns appearing
to be synchronous to some extent across the South Shetland Islands, South Orkney
Islands and South Georgia areas. This suggests a similarity in year-class strength of
krill across the Scotia Sea. Hence fishing companies face the problem of an overall
shortage of larger krill when krill are predominantly small-sized throughout the
south-west Atlantic.
There is some suggestion of long-term declines in krill abundance and recruitment in the South Shetland Islands area due to regional warming (Loeb et al., 1997).
CPUE of the Japanese fishery expressed in terms of catch per towing volume also
shows a long-term decreasing trend in this area (Kawaguchi et al., 1997). A possible
reason for the decreasing CPUE, is the use of different fishing strategies in response
to a demand for a higher product quality as well as a general decrease in krill density
in the area. Furthermore, in 1997±98, gravid females were caught in May (an
unusually late season for spawning) in the South Shetland Islands area (Kigami,
personal communication). The prevailing feeling among fishers, therefore, is that
Krill Harvesting
100
80
60
40
20
0
80/81
Size composition (%)
Size composition (%)
(a) SS
100
80
60
40
20
0
82/83
84/85
86/87
88/89
90/91
92/93
94/95
96/97
82/83
84/85
86/87
88/89
90/91
92/93
94/95
96/97
91
93
95
97
90/91
92/93
94/95
Size composition (%)
(b) SO
243
(c) SG
100
80
60
40
20
0
Size composition (%)
(d) WL
100
80
60
40
20
0
78/79
80/81
82/83
Size composition (%)
(e) KL
100
80
60
40
20
0
78/79
84/85
86/87
Body length class (mm)
>50
45–50
40–45
35–40
80/81
<35
82/83
Fishing season
Fig. 9.10 Interannual variations in body length composition of Antarctic krill for each fishing
area. SS: South Shetland Islands; SO: South Orkney Islands; SG: South Georgia; WL: Wilkes Land;
KL: Kemp Land.
244
Krill: Biology, Ecology and Fisheries
regional warming may well be affecting krill abundance and their biological characteristics in the south-west Atlantic.
9.3
Japanese northeastern coastal waters (Euphausia pacifica)
History of the fishery
About 100 years ago, dip net fishing, which later developed into the use of a bowmounted trawl, operated in Sanriku (Iwate and Miyagi Prefectures) coastal
waters for sand lance (Ammodytes personatus). The same method was applied to
E. pacifica in the mid-1940s by fishermen off Oshika Peninsula, Miyagi Prefecture (Odate, 1991; Fig. 9.11). Increasing bait requirements for sport fishing in
the late 1960s caused the E. pacifica fishery to expand along the northern and
southern coasts of Miyagi Prefecture. A fishery was subsequently established off
Ibaraki Prefecture in 1972, although large amounts of krill were not landed until
1977 (Fig. 9.12), when a coastal branch of the Oyashio Current extended far to
the south. Shortly thereafter in 1974, Fukushima Prefecture began a krill fishery
to be followed by the Iwate Prefecture in 1975 (Odate, 1991). In the 1980s,
increasing food demands for sea bream culture, and the introduction of highspeed fishing boats and effective fishing gear, such as fish pumps, echo-sounders,
sonar and larger net haulers, resulted in an increased catch rate (Kuroda, 1994).
With recent poor catches in other fisheries, such as that for sand lance, due to
overfishing, the krill fishery has developed into an important coastal water
fishery, with an average annual catch of around 60 000 t over the last 14 years
(Fig. 9.12).
Fishing strategy
Fishermen engaged in coastal fisheries attempt to maintain economic stability by
participating in diverse fishery operations throughout the year such as boat seines
for sand lance and juvenile sardine, squid jigging, gill nets for salmon and aquaculture for sea mustard. Coastal krill fishing was initially conducted during the offseason for the other species (Kuroda & Kotani, 1994), the importance of the fishery
being much greater in the Sanriku coastal area than the Joban (Fukushima and
Ibaraki Prefectures) coastal area, since catches were more stable in the former
locality.
There has been little market competition between Antarctic krill (E. superba)
and E. pacifica products, because they are sold for different purposes. For example,
in the case of sport fishing, E. pacifica is smaller and cheaper and used as chumbait,
whereas E. superba are mostly used as individual lures (Kigami, personal communication). Compared with E. superba, E. pacifica show less variability in body size
and feeding condition, resulting in the product of this species being graded only by
Krill Harvesting
Fig. 9.11
245
Map of northeastern Japan showing Euphausia pacifica fishing prefectures.
body colour. In contrast to E. superba, white-coloured E. pacifica are cheaper than
red ones by 10±15%. Fishermen, therefore, generally have less concern over krill
quality needing only to regulate catches in collaboration with their counterparts in
adjacent prefectures so as to keep the price high.
246
Krill: Biology, Ecology and Fisheries
120
Catch (1000 tonnes)
100
80
60
40
20
Iwate
Miyagi
Fukushima
1996
1994
1992
1990
1988
1986
1984
1982
1980
1978
1976
1974
1972
1970
1968
1966
1964
1962
1960
0
Ibaraki
Fig. 9.12 Annual catches of Euphausia pacifica in northeastern Japan. Data from Anon. (1998a).
Fishing fleet organisation
Only small boats of less than 20 t GRT are engaged in the fishery. Boat seines are
currently used in all prefectures, including Miyagi Prefecture, where they have been
used alongside bow-mounted trawls since 1991. The number of boats licensed for
the krill fishery varies from 300 to 500 in each prefecture, although not all operate
currently. The average number operating per day was 40±100 in each prefecture
from 1986 to 1993 (Kuroda & Kotani, 1994) with each fishing vessel operating more
or less individually.
Fishing operations are restricted to daytime for two reasons. Firstly, because ports
are only open during the daytime (i.e. from dawn to c. 15:00 h in the Sanriku coastal
area and from dawn to dusk in the Joban coastal area); and secondly, because fishing
vessels are not equipped with freezing facilities, and so they must return within a day
for onshore processing and freezing of their catch (Odate, 1991; Sawadate, 1993).
Characteristics of the fishing grounds
The use of small boats and single day operations has restricted the fishing areas to
nearshore waters. Fishing conditions are highly dependent upon the strength of the
cold Oyashio Current, the fishing grounds being formed near the front between the
coastal branch of the current (<58C) and coastal waters with an optimal surface
Krill Harvesting
247
water temperature of 78 to 98C (Odate, 1991). The onset of the fishing season is,
therefore, determined by the nearshore occurrence of optimal surface waters; such
information is available from the Japan Fisheries Information Service Center, which
provides sea-surface temperature data recorded by NOAA-9 AVHRR (Sawadate,
1993). Fishers also use local determinants, such as occurrence of certain seabirds
and marine mammals in coastal waters and the presence of E. pacifica in the
stomachs of demersal fish landed at the ports (Kotani, 1992). Individual vessels
maintain communication with each other and spread out to search for harvestable
concentrations (Sawadate, 1993). The fishing grounds thus exploited correspond to
the continental shelf (<200 m), within 10±20 nautical miles from the shore (Kuroda
& Kotani, 1994). It is yet to be determined whether or not fishable aggregations of
krill exist in the offshore regions (Taki, personal communication).
In the Sanriku coastal area, the krill fishing grounds tend to be stable since the
optimal water temperature occurs regularly, whereas in the Joban coastal area, it is
unstable since the optimal water temperature does not occur in years when the
Oyashio current is weak (Fig. 9.13). For example, the lack of fishing grounds in the
Joban coastal area in 1997 was related to a northward anomaly of the southern limit
of the first Oyashio current branch from the mean inter-annual value. By contrast,
fishing grounds established in the Joban coastal area in 1993 corresponded to the
southward extension of the first Oyashio current branch. As shown in Fig. 9.14, the
average daily catch rate expressed as tonnes/vessel for boat seines was high (> 5 t)
every year in the Sanriku coastal area, compared with substantial inter-annual
variability (0±4 t) in the Joban coastal area. Poor catches usually occur in years
characterised by the northward anomaly of the southern limit of the Oyashio
Current, such as occurred in 1997 (Fig. 9.13).
The fishing season also varies from area to area and from year to year, usually
occurring from February to April/May in the Sanriku coastal area, compared with
considerable inter-annual variation in the Joban coastal area. In the latter area, the
fishing season usually commenced in February in the mid 1980s, but often became as
late as May in the 1990s (Kuroda & Kotani, 1994).
Detection and catching of krill swarms
Swarms are detected using echo-sounders, sonar, and direct visual sightings of
surface aggregations and feeding seabirds. In the early days, when fishing boats were
not equipped with either echo-sounder or sonar, only surface swarms could be
fished. These were detected by the reddish colour of the sea surface or else the
presence of flocks of feeding seabirds. The latter provided an especially useful cue
for locating surface aggregations, owing to the greater distance from which the
observation could be made (Odate, 1991). However, with the introduction of echosounders, aggregations in mid- and deep-water were able to be detected (Suzuki,
1986). This opened the way for mid-water and bentho±pelagic aggregations to be
fished (Nakamura, 1992). Decreasing use of sighting cues in the 1990s is attributed
Fig. 9.13 Locations of Euphausia pacifica fishing grounds (black area) in relation to sea-surface temperatures in 1993 (cooler
year) and 1997 (warmer year) off northeastern Japan. Iw: Iwate; Mi: Miyagi; Fu: Fukushima; Ib: Ibaraki. Areas of sea temperature from 7±98C are shaded. Modified from `Quick reports on fishing and oceanographic conditions' from Japan Fisheries
Information Service Center.
248
Krill: Biology, Ecology and Fisheries
Krill Harvesting
249
Fig. 9.14 Interannual variations in average daily catch of Euphausia pacifica (tonnes/vessel) by
boat seine in each prefecture off northeastern Japan. Data from Anon. (1992; 1993; 1994; 1995a;
1996; 1998a.)
also to the decreasing occurrence of surface swarms due to increasingly warm surface water temperatures (Odate, 1992).
In Japanese coastal waters, not only E. pacifica but also finfish such as sand lance
can be predominant. Acoustic identification of species is thought to be possible
using characteristics in echo traces. At a speed of 10±12 knots, echo traces of krill
are either round or layered, while those of sand lances are vertical and needle-like
(Ebisawa, 1995). Discrimination between krill and walleye pollock (Theregra
chalcogramma) is also possible using characteristic differences in SV (volume
backscattering strength) between two frequencies (38 kHz and 120 kHz) (Miyashita
et al., 1997). The SV difference is large for krill, but small for walleye pollock.
Of the two krill fishing methods employed, bow-mounted trawls, which have a net
opening of approximately 75 m2, only catch aggregations within 8 m of the surface.
In this method, when an aggregation is sighted, the boat approaches and two poles
slide forward like probing antennae (Fig. 9.15a). The booms tilt and plunge, causing
the net to spread open beneath the bow. The boat pushes the net slowly through the
aggregation, engulfing the krill (Kodama, 1995a). The average catch per haul is
approximately 300 kg (Odate, personal communication).
Boat seines, the other fishing method, can catch aggregations as deep as 150 m.
When an aggregation is detected, a buoy fixed to the free end of a rope is thrown
250
Krill: Biology, Ecology and Fisheries
Fig. 9.15 Methods for catching Euphausia pacifica: by bow-mounted trawl (a), and boat seine (b)
off northeastern Japan. Redrawn from Kodama (1995b).
Krill Harvesting
251
overboard, and the wing net, 100 m long by 50 m high, pocket net and second wing
net are streamed out to surround the krill aggregation (Fig. 9.15b). After the buoy
has been retrieved, both ropes are hauled whilst ensuring that the vessel orientates
downwind and down current (Kodama, 1995a). The average catch per haul is
greater than 1000 kg (Sawadate, 1993).
Bow-mounted trawls, which are now only used in the Miyagi Prefecture, are in the
process of being replaced by boat seines, due to the former's limitations in trawling
depth and low CPUE. The proportion of catches by bow-mounted trawls relative to
the total catches in this Prefecture has been negligible for the past two years, 1996
and 1997.
In the Sanriku coastal area, krill are not distributed in deeper water due to the
strong upwelling of very cool water of less than 48C, well below their optimal
temperature. Arising from this, the fishing depths tend to be shallower than 50 m. In
the Joban coastal area, on the other hand, as the surface water temperature
increases towards the end of the fishing season, surface aggregations of E. pacifica
descend to cooler mid- and deep-water layers and fishing depths can be deeper than
100 m (Odate, 1992). Nakamura (1992) speculated that in the Joban coastal area,
when the water is cold (58C) and unstratified, krill aggregations occur throughout
the water column, whereas mid-water and benthopelagic aggregations occur when
the water column is stratified, with water warmer than 108C at the surface and an
extensive, cold deep layer where the temperature varies from 68 to 98C.
In both types of fishing operation, boat seine and bow-mounted trawl, fish pumps
are used to transfer krill from the cod end to plastic containers, each of which can
hold about 30 kg of krill. The catch is regulated by controlling the number of these
containers allowed on board each vessel.
Primary concerns of the fishers
For fishers, especially in the Joban coastal area, besides the limited market for krill
products, the most important concerns related to the krill fishery are the ability to
predict fishing conditions, the length of the fishing season and the area of occurrence of the fishery.
Kotani (1992) undertook a correlation analysis between krill catches in various
coastal areas and the southernmost latitude of the coastal branch of the Oyashio
Current. A significant correlation was found between the annual catch in the Ibaraki
Prefecture, where fishing conditions were the least stable, and the southernmost
latitude of the coastal branch of the Oyashio current during the fishing season (r =
70.822, p < 0.05, n = 8). In other words, the greater the southward extension of the
coastal branch of the Oyashio Current, the higher the catch in this prefecture. Thus,
the southernmost latitude of the coastal branch of the Oyashio Current during the
fishing season is crucial to predictions of fishing conditions. Kotani (1992) also
found a close correlation between the southernmost latitude of the Oyashio Current
during the fishing season and that in the preceding January (r = 0.71, p < 0.01, n =
252
Krill: Biology, Ecology and Fisheries
13). Based on this, a prediction of krill catch for a fishing season, may be possible a
few months prior to the beginning of that season.
The first fishing day of the year is thought to be best predicted according to the
distribution of water masses. Assuming that the optimal temperature for E. pacifica
aggregations is 58 to 108C, Kotani (1992) found a significant correlation between the
distance from the Otsu fishing ground in Ibaraki Prefecture to the water mass with a
temperature of 108C at 100 m depth and the number of days from 25 January to the
date when fishing commenced in that year for each prefecture (r = 0.917, p < 0.01,
n = 7). In other words, the closer the water mass with a temperature of 108C at a
depth of 100 m to the Otsu fishing ground, the greater the likelihood of fishing
commencing in late January.
The southernmost latitude of the coastal branch of the Oyashio Current
showed long-term inter-annual changes (Kodama, 1995b). From 1948 to 1973, a
warm water period, the current tended to shift northward, but from 1974 to 1987,
a cool water period, it tended to shift southward. From 1988 to the present,
another warm water period, it has frequently shifted northward again. Historical
catches of E. pacifica seem to be related to these long-term changes in the
southern limit of the Oyashio Current. In the early- to mid-1980s, Ibaraki Prefecture expanded its catch to around 50% of the total catch, but after 1988, Iwate
and Miyagi Prefectures became the main fishing prefectures (Fig. 9.12). Since the
krill fishery is licensed by prefectural governors and hence fishing activity permitted only within prefectural boundaries, drastic changes in the location of fishing grounds across boundaries impact greatly on the overall management of these
coastal fisheries overall. Therefore, long-term prediction of fishing conditions is
an important consideration for management.
To date, catch and CPUE data are used to evaluate fishing conditions, but they
are unlikely to be of much use as indicators of krill abundance because both catch
and fishing effort are regulated by the industry. For more reliable predictions of
fishing conditions (including location of fishery and length of the fishing season),
fishery-independent monitoring such as periodic surveys of abundance and distribution of krill in relation to oceanographic conditions, is necessary.
9.4
Canadian west coast waters (Euphausia pacifica)
History of the fishery
A trial commercial fishery for E. pacifica began in about 1970 in the Strait of
Georgia and east of Vancouver Island (Fig. 9.16). Over the ensuing years a market
for frozen and freeze-dried euphausiids developed, with the products being used for
aquaculture and the aquarium pet food industry (Haig-Brown, 1994). Until 1985,
annual catches were less than 200 t, with fishing concentrated initially in Saanich
Inlet. Subsequent developments extended the fishery into Howe Sound and most
recently in Jervis Inlet (Anon, 1995b). In 1990, the largest catch, totalling 530 t, was
Krill Harvesting
253
Fig. 9.16 Location of Euphausia pacifica fishery off the west coast of Canada. Redrawn from
Nicol & Endo (1997). With permission from the Food and Agriculture Organisation of the United
Nations.
reported (Fig. 9.17a). In 1992, catches totalling 381 t were reported, with catch rates
during November being exceptionally high as fishers found extremely dense
aggregations of krill. As a result of this glut a large quantity of krill was landed over
a short period of time which exceeded the processing capacity. Consequently, some
poor quality and spoiled krill reached the market, discouraging buyers. Subsequent
buyer resistance in 1993 resulted in a very much decreased catch that year, totalling
only 53 t. However, krill harvests during 1994 recovered to the previous high levels
due to renewed market interest (Nicol & Endo, 1997).
Fishing strategy
This fishery is conducted during the out-of-season period for the salmon fishery.
The market is limited, with the majority of the products being frozen for export to
254
Krill: Biology, Ecology and Fisheries
600
(a)
500
Catch (tonnes)
400
300
200
100
0
1983
1985
1987
1989
1991
1993
1991
1993
1995
Year
1200
(b)
CPUE (kg/hr)
1000
800
600
400
200
0
1983
1985
1987
1989
Year
Fig. 9.17 (a) Annual catches of Euphausia pacifica off the west coast of Canada. From Nicol &
Endo (1997). (b) Interannual variations in average catch per towing time for Euphausia pacifica
fishery off the west coast of Canada. Data from Nicol & Endo (1997).
Krill Harvesting
255
the US where it is used in the aquaculture and pet food industries. As in the case of
the krill fishery in Japanese northeastern coastal waters, fishermen regulate catches
in order to maintain a high price.
Fleet organisation
Two types of vessels participate in the fishery; smaller freezer vessels, whose catches
are limited by their low freezing capacity (5±6 t/day) and larger vessels which land
greater quantities of krill for onshore processing and freezing (Haig-Brown, 1994).
The catch must be frozen within 24 h so as to avoid significant deterioration in
product quality. The fishing season can be as short as 20 days (actual fishing days),
with individual vessels landing as little as 32 t in a season. The larger vessels, which
use seine nets and have no onboard freezing capacity, are usually out-of-season
salmon fishing boats, their presence in the krill fishery being usually dependent
upon the success of the salmon fishery. If there has been a bad salmon catch, krill are
fished to boost revenue (Nicol & Endo, 1997).
The number of licences issued for this fishery increased annually from 7 in 1983 to
56 in 1990, but decreased to 45 in 1991 (Table 9.2). In 1992 licences were limited to
25 vessels following industry advice and because the annual quota was being landed
by the vessels already operating. Despite the number of licences issued, the number
of vessels reporting landings has been varied from only one in 1993, to a maximum
of 17 in 1990. Seven vessels reported euphausiid landings in 1995 and ten in 1996
(Anon, 1998b).
Table 9.2 Numbers of licences issued and fishing vessels and reported landings for the Euphausia
pacifica fishery off the west coast of Canada.
Year
Number of licences
issued
Number of vessels
fishing
Fishing
days
Quota
(t)
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
7
8
5
11
18
24
45
56
45
25
18
18
18
2
4
2
2
3
4
15
17
14
10
1
6
7
50
67
65
67
36
11
225
255
152
87
29
122
163
500
500
500
525
525
785
500
500
500
500
500
500
500
Data from Nicol & Endo (1997)
256
Krill: Biology, Ecology and Fisheries
Characteristics of the fishing grounds
Currently, most of the commercial krill harvest has come from either the mouth of
Jervis Inlet or the adjacent Malaspina Strait within the Strait of Georgia. Because
Jervis Inlet (c. 316 km) is mostly enclosed, short-term changes in krill abundance
due to physical transport are unlikely (Romaine et al., 1995). The commercial
fishery is limited to the winter months, from November to March each year, to
minimise the incidental catch of larval and juvenile fish, and shellfish. Not only does
this ensure the least impact on fish and shellfish resources, but also by minimising
by-catch, it improves the quality of krill products (Nicol & Endo, 1997).
Detection and catching of krill swarms
As in other fisheries, krill are located by echo-sounders. The nets that are used have
a mouth area of around 80 m2 the trawl mouth being kept open by a beam and being
buoyed so as to keep it from flipping over when the ship turns, while weights on the
footrope maintain the net shape. Fishing is carried out, at speeds of less than one
knot, close to the surface ± often less than 20 m deep at night and on moonless nights
when the krill rise to the surface forming layers less than 10 m in vertical extent.
Fishing depth is determined largely by the speed of the net through the water (HaigBrown, 1994).
Catch per unit effort (CPUE in kg/h) reported on harvest logs has remained
relatively constant since 1986 at approximately 300 to 400 kg/h (Fig. 9.17b). During
the year of greatest effort (1988) CPUE was only 255 kg/h. The highest CPUE (1153
kg/h) was reported in 1992, when fishers reported fishing extremely dense concentrations of krill in Malaspina Strait during November. The CPUE for 1993 was
565 kg/h (Fig. 9.17b). (Nicol & Endo, 1997).
9.5
Comparisons of the three krill fisheries
In Table 9.3 the main features of the three current krill fisheries are summarised.
Fishing grounds in the Antarctic and off western Canada are associated with
topographical features such as islands, continental shelves, inlets and straits, while
those in Japanese northeastern coastal waters are associated with the frontal area
between the Oyashio Current and coastal waters. In the latter case, the fishing
grounds are also restricted to the continental shelf within 10±20 n.miles from the
shore, which means that they tend to form in a similar area each year. Thus, any krill
operations tend to be concentrated within certain restricted areas. This should be
noted for any assessment of the impact of fishing on local krill abundance and
subsequent effect on dependent species.
In the Antarctic, the USSR formerly harvested krill throughout the year, but the
Japanese Antarctic krill fishery, which takes most of the current catch, operates only
Krill Harvesting
257
Table 9.3 Comparative information on current krill fisheries.
Antarctic Ocean
Off northeastern Japan
Off western Canada
Target species
Euphausia superba
E. pacifica
E. pacifica
Fishing ground
The vicinity of
islands and
continental slope
Front between the
Oyashio current and
coastal waters
Inlet and strait
Fishing season
December±August
February±May
November±March
Type of fishing boat
Stern trawlers
(3000±4000 t)
Small boats (< 20 t)
Small boats
No. of fishing boats
< 10
> 300
< 20
Fishing method
Mid-water trawling
Boat seine
Surface trawling
Fishing time
Day and night
Daytime
Moonless night
Annual catch
c. 100 000 t
c. 60 000 t
c. 300±400 t
CPUE
c. 100 tonnes/vessel
day
c. 10 tonnes/haul
c. 5 tonnes/vessel/
day
c. 1 tonne/haul
c. 3 tonnes/vessel/
day
c. 0.5 tonne/haul
Favourite krill
quality
Larger white krill
are preferred
There is no quality aspect
except body colour
There is no quality
aspect
Fisherman's
concern
Harvesting of
preferred krill
Prediction of fishing
condition, length of
fishing season and
location of fishing
grounds
Availability of haulby-haul data
Yes
No
Yes
Current human
consumption
23% of total catch
6±7%
Negligible
from December to August. This allows alternative fisheries such as for squid and fin
fish to operate, and at the same time allows vacation time for crews. The fishing
season for E. pacifica in northeastern Japan is from February to May, when that
species is available in the coastal waters. In western Canada, however, fishing is
limited to the period from November to March, when incidental catches of larval
fish and shellfish are likely to be at a minimum. More than 300 fishing vessels are
involved in the krill fishery off northeastern Japan, which makes the fishery very
important to the local coastal region. Accordingly, the North Pacific Krill Resources
Research Group was established in 1992 to conduct demographic studies, undertake
biomass estimates and develop methods for predicting fishing conditions.
The time of day when fishing operations take place for E. pacifica is completely
different in northeastern Japan when compared with western Canada. In the former
fishing takes place during the daytime when ports are open for onshore processing
258
Krill: Biology, Ecology and Fisheries
and freezing, whereas in the latter fishing is only at night when krill aggregations are
near the surface. Annual catches are restricted by market limitations in all the
fisheries. Since boats fishing in the Antarctic are larger in size compared with those
in the other two fisheries, CPUEs (tonnes/vessel/day and tonnes/haul) in the former
are at least one order of magnitude greater than elsewhere.
E. superba is a larger species, growing to a maximum size of 65 mm (Baker et al.,
1990) and is thus subject to substantial size variability at harvest. Furthermore, other
quality aspects, degree of `greenness' and body colour show great variability (Plate 2,
facing p. 182), with a resulting emphasis on making catches which will result in `high
quality' products being at a premium. E. pacifica, on the other hand, reaches a length
of 25 mm (Baker et al., 1990), and has less variability in body size when harvested
compared with E. superba. Variability in feeding conditions is also less distinct.
Hence, body colour of E. pacifica is the only quality of concern to the fishers.
In the Antarctic krill fishery, detailed fishing data, i.e. haul-by-haul catch and
effort data, have been collected on a continuing basis, thus providing much useful
information on spatial and temporal changes in catches, CPUE and size of krill
(Endo & Ichii, 1989; Ichii, 1990; Marin et al., 1991; Fedulov et al., 1996; Kawaguchi et
al., 1996; 1997; Murphy et al., 1997; Trathan et al., 1998). For example, Japanese
haul-by-haul data include the following:
(1)
(2)
(3)
Trawling information ± date, position trawling started, time trawling started,
trawling time, trawling depth, direction and speed of trawling.
Environment ± weather, wind direction and force, atmospheric pressure, air
and sea surface temperatures, and pack-ice conditions.
Catch record ± weight of each commercial size category, weight of discard, and
total catch.
Biological data include the body length of 50 individuals of krill randomly sampled from one haul per day. Currently, measurements of `greenness' and salp bycatch information have been included in the haul-by-haul data (see Kawaguchi et al.,
1998). Comparable haul-by-haul data are collected in the Canadian krill fishery
(Anon, 1998b). Analyses of these should provide more details on the fishing
activities for and biological aspects of E. pacifica.
Acknowledgements
I wish to thank K. Taki and T. Ogishima, Tohoku National Fisheries Research
Institute, K.Odate, and Y. Kotani, Nansei National Fisheries Research Institute, for
their valuable information on the E. pacifica fishery off northeastern Japan. I am
also grateful to Gregar Saxby, Biozyme Systems Inc., for information on the E.
pacifica fishery in Canadian west coast waters. I also thank M. Kigami, Maruha Co.,
Ltd., and S. Kaneda, Nissui Co., Ltd. for information on current E. superba fishery.
Krill Harvesting
259
References
Anon. (1992) Report of the Research Meeting on North Pacific Krill Resources 1, Tohoku
National Fisheries Research Institute, Fisheries Agency (in Japanese).
Anon. (1993) Report of the Research Meeting on North Pacific Krill Resources 2, Tohoku
National Fisheries Research Institute, Fisheries Agency (in Japanese).
Anon. (1994) Report of the Research Meeting on North Pacific Krill Resources 3, Tohoku
National Fisheries Research Institute, Fisheries Agency (in Japanese).
Anon. (1995a) Report of the Research Meeting on North Pacific Krill Resources 4, Tohoku
National Fisheries Research Institute, Fisheries Agency (in Japanese).
Anon. (1995b) Pacific Region, 1995 management plan. Plankton ± Euphausiids: Fisheries and
Oceans, Canada.
Anon. (1996) Report of the Research Meeting on North Pacific Krill Resources 5, Tohoku
National Fisheries Research Institute, Fisheries Agency (in Japanese).
Anon. (1998a) Report of the Research Meeting on North Pacific Krill Resources 6, Tohoku
National Fisheries Research Institute, Fisheries Agency (in Japanese).
Anon. (1998b) Pacific Region, 1998 management plan. Plankton ± Euphausiids: Fisheries and
Oceans, Canada.
Baker, A. de C., Boden, B. P. and Brinton, E. (1990) A Practical Guide to the Euphausiids of
the World. Natural History Museum Publications, London.
Butterworth, D.S. & Miller, D.G.M. (1987) A note on relating Antarctic krill catch-per-uniteffort measures to abundance trends. S. Afr. J. Antarctic Res. 17, 112±16.
CCAMLR (1990a) Statistical Bulletin, Vol. 1 (1970±1979). CCAMLR, Hobart, Australia
(CCAMLR-SB/90/1).
CCAMLR (1990b) Statistical Bulletin, Vol. 2 (1980±1989). CCAMLR, Hobart, Australia
(CCAMLR-SB/90/2).
CCAMLR (1997) Statistical Bulletin, Vol. 9 (1987±1996). CCAMLR, Hobart, Australia
(CCAMLR-SB/97/9).
Dolzhenkov, V.N., Lubimova, T.G., Makarov, R.R., Parfenovich, S.S. & Spiridonov V.A.
(1988) Some peculiarities of the USSR krill fishery and possibilities to use fishery statistics
in studies of krill biology and stocks. In Selected Scientific Papers, 1988 (SC-CAMLR-SSP/
5), Vol. Part I, pp. 237-252. CCAMLR, Hobart, Australia.
Ebisawa, Y. (1995) Characteristics of echo-traces of krill. In Manual for species identification,
measurements of body size and sampling of Euphausiid species off Sanriku and Joban areas.
(North Pacific Krill Resources Research Group, ed.). 32 pp. (in Japanese).
Eddie, G.C. (1977) The harvesting of krill. Southern Ocean Fisheries Survey Programme,
FAO Rome. GLO/SO/77/2.
Endo, Y. & Ichii, T. (1989) CPUEs, body length and greenness of Antarctic krill during 1987/
88 season in the fishing ground north of Livingston Island. In Selected Scientific Papers,
1989 (SC-CAMLR-SSP/6), pp. 323±38, CCAMLR, Hobart, Australia.
Everson, I. (1978) Antarctic fisheries. Polar Record 19, 233±51.
Everson, I. (1988) Can we satisfactorily estimate variation in krill abundance? In Antarctic
Ocean and Resources Variability (D. Sahrhage, ed.), pp. 199±208. Springer-Verlag, Berlin.
Everson, I. & Goss, C. (1991) Krill fishing activity in the southwest Atlantic. Antarctic Sci. 3,
351±58.
Fedulov P.P., Murphy, E. & Shulgovsky, K.E. (1996) Environment-krill relations in the South
Georgia marine ecosystem. CCAMLR Sci. 3, 13±30.
260
Krill: Biology, Ecology and Fisheries
Fisher, L.R., Kon, S.K. & Thompson, S.Y. (1953) Vitamin A and carotenoids in some
Mediterranean crustacea with a note on the swarming of Meganyctiphanes. Bull. Inst.
Oceanogr. Monaco 1012, 1±19.
Haig-Brown, A. (1994) Going for the krill. In National Fishermen, May 1994, pp. 18±19.
Ichii, T. (1987) Observations of fishing operations on a krill trawler and distributional
behavior of krill off Wilkes Land during the 1985/86 season. In Selected Scientific Papers,
1987 (SC-CAMLR-SSP/4), pp. 335±63. CCAMLR, Hobart, Australia.
Ichii, T. (1990) Distribution of Antarctic krill concentrations exploited by Japanese krill
trawlers and minke whales. In Proceedings of NIPR Symposium on Polar Biology 3, 36±56.
Ichii, T., Katayama, K., Obitsu, N., Ishii, H. & Naganobu, M. (1998) Occurrence of Antarctic
krill (Euphausia superba) concentrations in the vicinity of the South Shetland Islands:
relationship to environmental parameters. Deep-Sea Res. I 45, 1235±62.
Ikeda, T. & Dixon, P. (1982) Body shrinkage: a possible over-wintering strategy of the
Antarctic krill (Euphausia superba). J. Exp. Mar. Biol. Ecol 62, 143±51.
Kawaguchi, S., Ichii, T. & Naganobu, M. (1996) CPUE and net towing depth and body length
of krill during the winter operation of Japanese krill fishery around South Georgia.
Working paper for CCAMLR; Working Group on Ecosystem Monitoring and Management
(WG-EMM), 96/51. CCAMLR, Hobart, Australia.
Kawaguchi, S., Ichii, T. & Naganobu, M. (1997) Catch per unit effort and proportional
recruitment indices from Japanese krill fishery data in Subarea 48.1. CCAMLR Sci. 4, 47±
63.
Kawaguchi, S., de la Mare, W.K., Ichii, T. & Naganobu, M. (1998) Do krill and salp compete ?
Contrary evidence from the krill fisheries. CCAMLR Sci. 5, 205±216.
Kock, K.-H. (1992) Antarctic Fish and Fisheries. Cambridge University Press, Cambridge.
Kock, K.-H. (1994) Fishing and conservation in southern waters. Polar Rec. 30, 3±22.
Kodama, J. (1995a) Current status and problems of prediction of fishing condition for
Euphausia pacifica in Miyagi Prefecture. Bull. Japanese Soc. Fish. Oceanogr. 59, 145±47.
Kodama, J. (1995b) Commercial krill fishing net. In Manual for species identification, measurements of body size and sampling of Euphausiid species off Sanriku and Joban areas.
(North Pacific Krill Resources Research Group, ed.), pp. 29±31 (in Japanese).
Kotani, Y. (1992) How to predict fishing conditions of Euphausia pacifica. Report of the
Research Meeting on North Pacific Krill Resources 1, 108±113 (in Japanese).
Kuroda, K. (1994) Euphausiid fishery in the Japanese waters and related research activities.
Gekkan Kaiyo 26, 203±209. (in Japanese).
Kuroda, K. & Kotani, Y. (1994) Euphausiid fishery and interannual variation of the fishing
conditions in the Sanriku-Joban coastal areas. Gekkan Kaiyo 26, 210±217 (in Japanese).
Laws, R.M. (1977) Seals and whales of the Southern Ocean. Phil. Trans. Roy. Soc., London
279, 81±96.
Loeb, V., Siegel, V., Holm-Hansen, O., Hewitt, R., Fraser, W. & Trivelpiece, W. (1997)
Effects of sea-ice extent and krill or salp dominance on the Antarctic food web. Nature 387,
897±900.
Marin, V.H., Mujica, A. & Eberhard, P. (1991) Chilean krill fishery: analysis of the 1991
season. In Selected Scientific Papers, 1991 (SC-CAMLR-SSP/8), pp. 273±87. CCAMLR,
Hobart, Australia.
Marr, J.W.S. (1962) The natural history and geography of the Antarctic krill (Euphausia
superba). Discovery Rep. 32, 33±464.
Krill Harvesting
261
Miyashita, K., Aoki, I., Seno, K., Taki, K. & Ogishima T. (1997) Acoustic identification of
isada krill Euphausia pacifica Hansen, off the Sanriku coast, north-eastern Japan. Fish.
Oceanogr. 6, 266±71.
Murphy, E.J., Trathan, P.N., Everson, I., Parkes, G. & Daunt, F. (1997) Krill fishing in the
Scotia Sea in relation to bathymetry, including the detailed distribution around South
Georgia. CCAMLR Sci. 4, 1±17.
Nakamura, T. (1992) Recent aspect of krill fishing grounds off Joban-Kashima area in
relation to warming tendency. Bull. Japanese Soc. Fish. Oceanogr. 56, 155±57.
Nicol, S. (1991) CCAMLR and its approaches to management of the krill fishery. Polar Rec.
27, 229±36.
Nicol, S. & Endo, Y. (1997) Krill fisheries of the world. FAO Fisheries Technical Paper 367, 1±
100.
Odate, K. (1991) Fishery biology of the krill, Euphausia pacifica, in the northeastern coasts of
Japan. Fishery Science Library, 40 (in Japanese).
Odate, K. (1992) History and present status of the krill (Euphausia pacifica) fishery. Report
of the Research Meeting on North Pacific Krill Resources 1, 72±82 (in Japanese).
Romaine, S.J., Mackas, D.L., Macaulay, M.C. & Saxby, D.J. (1995) Comparisons of repeat
acoustic surveys in Jervis Inlet, British Columbia, 1994±1995. In Harvesting krill: Ecological
impact, assessment, products and markets, Vol. 3(3) (T.J. Pitcher & R. Chuenpagdee, eds).
Vancouver: University of British Columbia Fisheries Centre Research Reports, pp. 48±50.
Sawadate, M. (1993) Current status and problem of the krill (Euphausia pacifica) fishery in
the vicinity of Otsuchi. Report of the Research Meeting on North Pacific Krill Resources 2,
87±88 (in Japanese).
SC-CAMLR (1992) Report of the Eleventh Meeting of the Scientific Committee. CCAMLR,
Hobart, Australia (SC-CAMLR-XI).
SC-CAMLR (1997) Report of the Sixteenth Meeting of the Scientific Committee. CCAMLR,
Hobart, Australia (SC-CAMLR-XVI).
Shimadzu, Y. (1984) A brief summary of Japanese fishing activity relating to Antarctic krill,
1972/73±1982/83. In Scientific Papers 1982±1984, pp. 439±71. CCAMLR, Hobart, Australia.
Siegel, V. (1987) Age and growth of Antarctic Euphausiacee (Crustacea) under natural
conditions. Marine Biology 96, 483±95.
Siegel, V. (1988) A concept of seasonal variation of krill (Euphausia superba) distribution and
abundance west of the Antarctic Peninsula. In Antarctic Ocean and Resources Variability
(D. Sahrhage, ed.), pp. 219±30. Springer, Berlin.
Siegel, V. & Loeb, V. (1995) Recruitment of Antarctic krill Euphausia superba and possible
causes for its variability. Mar. Ecol. Prog. Ser. 124, 45±56.
Siegel, V., de la Mare, W.K. & Loeb, V. (1997) Long-term monitoring of krill recruitment and
abundance indices in the Elephant Island area (Antarctic Peninsula). CCAMLR Sci. 4, 19±
36.
Suzuki, M. (1986) On the distribution of Euphausia pacifica in the southern Joban and
Kashima-nada areas. Bull. Tohoku Branch Japanese Soc. Fish. Sci. 37, 30±31 (in Japanese).
Trathan, P.N., Everson, I., Murphy, E.J. & Parkes, G. (1998) Analysis of trawl data from the
South Georgia krill fishery. CCAMLR Sci. 5, 9±30.
Chapter 10
Products Derived from Krill
Stephen Nicol, Ian Forster and John Spence
10.1
Introduction
Small-scale krill fisheries, which have provided sources of fishing bait and feed for
fish culture, have developed in a number of regions (Fisher et al., 1953; Mauchline &
Fisher, 1969). In some areas these fisheries have grown into larger-scale operations
(Nicol & Endo, 1997, 1999). Lately, a fishery for Antarctic krill (Euphausia superba)
has been carried out on a large scale (Miller, 1991). The Antarctic krill fishery
developed primarily because of the large size of the krill population and its apparent
ease of harvesting, but development has slowed recently because of the high cost of
fishing, and the lack of a suitable product, or products, with a reliable and effective
economic return (Bykowski, 1986). Because most of the northern hemisphere krill
fisheries were developed to supply localised markets for bait and for aquaculture
feed, there has been limited development of new products from northern fisheries,
until recently. In contrast, there has been a considerable effort devoted to producing
a range of products from the fishery for Antarctic krill, and in developing a market
for its products. The products of the Antarctic krill fishery have been reviewed a
number of times (Eddie, 1977; Everson, 1977; Grantham, 1977; Suzuki, 1981;
Budzinski et al., 1985; Suzuki & Shibata, 1990). We will concentrate primarily on
recent developments in fisheries products and their uses.
10.2
Constraints to using krill
Introduction
Krill are generally smaller in size than most other organisms that have been commercially harvested. Antarctic krill, with a maximum wet weight of approximately 2
g, is the largest euphausiid that has been harvested, while the smallest harvested krill
are of the order of 0.01±0.02 g (Nicol & Endo, 1997). Although other small species
of crustaceans have been commercially harvested, and there are large fisheries for
species such as sergestid shrimps (Parsons, 1972; Omori, 1978; Neal & Maris, 1985),
the small size of krill, and some of their biological characteristics, have made their
harvesting and processing particularly problematic.
262
Products Derived from Krill
263
Rapid spoiling
The digestive gland in the cephalothorax of krill contains powerful hydrolytic
enzymes, including proteases, carbohydrases, nucleases and phospholipases, which
begin to break down the body tissues immediately following death, particularly if
crushing has occurred (Bykowski, 1986). Because these enzymes cohabit, they are
mutually protected against their degrading effects, a property rare in nature
(Anheller et al., 1989). These enzymes are controlled by an inhibitor system which is
disabled on death, facilitating the rapid process of autolysis (Sjodahl et al. 1998).
The characterisation of these enzymes has been most complete for Antarctic krill
(Sjodahl et al., 1998), but there has been some research into the enzymes of
Meganyctiphanes norvegica (Peters et al., 1998) as well. Because of the different
trophic niches of species of krill, it is likely that they will possess different suites of
enzymes, but it appears that most species, certainly those that are currently harvested, all suffer from rapid spoiling as a result of autolytic processes (Bykowski,
1986). Although this property of krill has its drawbacks for commercialisation, it
also has been utilised in the development of commercial products (see later sections).
The level of digestive enzymes in krill is highest in those individuals that have
been actively feeding. These animals have a distinct greenness in the digestive gland,
which is associated with a `grassy' flavour in krill products manufactured for human
consumption (Bykowski, 1986). Actively feeding krill also contain more acids and
some ketones, alcohols and sulphur compounds than non-feeding krill, and the
presence of volatile compounds affects the odour of krill (Gajowiecki, 1995). For
these reasons, green krill are avoided by the Antarctic fishery and the catch is
graded on its quality by reference to its greenness (Ichii, Chapter 9). Charts for this
purpose have been produced (CCAMLR 1993; Plate 2, facing p. 182).
Krill proteins have a relatively high level of solubility, when compared to fish
proteins, and this solubility increases with the degree of autoproteolysis (Kolokowski, 1989). This can present challenges in temporary storage prior to processing
(Bykowski, 1986), when loss of soluble fractions can occur. The high solubility of
krill proteins also has some advantages for producing certain types of end product,
however (see hydrolysates in section on aquaculture feed additives).
The lipids of E. superba are subject to change during refrigerated storage
(Kolakowska, 1988), with the critical factors being the time between capture and
freezing and the temperature of freezing. Free fatty acids increase markedly following death, and rapid deep freezing is necessary to maintain product quality.
Bacteria
The bacterial flora of Antarctic krill have been described, and their activity in the
gut may contribute significantly to spoilage (Donachie & Zdanowski, 1998), but
their abundance and role as krill symbionts is uncertain (Fevolden, 1981; Virtue et
264
Krill: Biology, Ecology and Fisheries
al., 1997). They do seem, however, to play a smaller role in the post-capture
breakdown of krill than do auto digestive processes (Rakusa-Suszczewski &
Zadanowski, 1980). Bacteria have also been shown to have a digestive function in
M. norvegica, but the spoilage effect of bacteria on species of krill harvested at
higher temperatures is unknown (Donachie et al., 1995).
Parasites
Krill may be intermediate hosts of parasites, which can be passed on to other
organisms that ingest them; this may be of particular importance when uncooked or
unprocessed krill are used in aquaculture feeds. The issue of biosecurity has been
viewed as the greatest threat to shrimp aquaculture, and there is particular sensitivity concerning the introduction of viruses in feed produced from other crustaceans (Lotz, 1997). Nothing is known about the viral infections of krill.
There is evidence that M. norvegica and Thysanoessa raschii are important
intermediate hosts of the helminth Anisakis simplex (Hays et al., 1998). North
Pacific euphausiids have a low level of occurrence of larval digeneans, cestodes,
nematodes and acanthocephalans (Kagei, 1985), and such parasites were looked for
but not found in Antarctic krill (Kagei et al., 1978). Other symbionts of krill with less
harmful potential have also been reported (Nemoto, 1970; Kulka & Corey, 1984;
Nicol, 1984; Rakusa-Suszczewski & Filcek, 1988), but reports on studies examining
parasitism in krill are rare.
Fluoride
The high fluoride content of the exoskeleton of krill was first indicated by research
into E. superba (Soevik & Breakkan, 1979), but all other species of krill so far
examined have been found to have similarly high levels (Sands et al., 1998). It seems
likely that high exoskeleton fluoride concentration is a general feature of
euphausiids and that fisheries on krill will have to take this feature into account
when assessing potential products.
The fluoride in krill is localised in the exoskeleton, where it can reach concentrations of 3500 mg F g±1 dry weight (Virtue et al., 1995), but concentrations in the
muscle and other internal tissues appear to be less than 100 mg F g±1 dry weight
(Sands et al., 1998). Once krill die, however, there is very rapid leaching of the
fluoride from the exoskeleton into the tissue (Adelung et al., 1987), even if frozen to
7208C (Christians & Leinemann, 1983). Freezing to temperatures lower than
7308C is necessary to prevent migration of fluoride from the shell into the muscle
tissue. Rapid peeling appears to be the most efficient way of separating the fluoriderich shell from the flesh, although boiling also fixes fluoride in the shell (Bykowski,
1986). The chemical form of fluoride in the shell is unknown but there are techniques which can produce a low fluoride (5±21 mg F g±1 dry weight) krill paste or krill
protein concentrate by washing with organic acid or water (Tenuta-Filho, 1993).
Products Derived from Krill
265
Vertebrates that are fed on krill tend to accumulate fluoride to deleterious levels
in their bones and tissues (Krasowska, 1989) and the levels of this element in krill
meal are as much as four times the allowable levels in feed for the European
Community (Bykowski, 1986). Those animals with a natural diet that contains krill,
however, appear to be able to maintain low tissue fluoride levels and tolerate high
bone fluoride levels (Schneppenheim, 1980; OehlenschlaÈger & Manthey, 1982).
Many fish species naturally eat krill and do not appear to accumulate fluoride in
their tissues or bones, consequently, whole krill can be used as an aquaculture feed
for many species without long-term accumulation of tissue fluoride (Grave, 1981).
Despite many feeding trials on the use of krill meal for a variety of domestic
animals, it seems that the high fluoride level of most krill meal, and its high cost,
have prevented further developments in this area (Kotarbinska & Grosyzk, 1977;
OehlenschlaÈger, 1979; Bykowski, 1986).
10.3
Current uses of krill
Introduction
There has been considerable effort expended in developing Antarctic krill products
for human consumption, but, most of the krill catch has been used for domestic
animal feed, and, particularly in recent years, for aquaculture feed. The Japanese
Antarctic krill fishery, which takes most of the current catch, produces four types of
product: fresh frozen (46% of the catch), boiled-frozen (10% of the catch), peeled
krill meat (10% of the catch) and meal (34% of the catch). These products are used
for aquaculture and aquarium feed (43% of the catch), for sport fishing bait (~ 45%
of the catch) and for human consumption (~12% of the catch) (1999 figures, T. Ichii,
Japan National Research Institute of Far Seas Fisheries, pers. comm.).
E. pacifica caught off Japan is used for sport fishing (~ 50% of the catch), feed for
fish culture (particularly as a reddening agent), and a small amount is used for
human consumption (Kuroda, 1994). Most of the E. pacifica from the Canadian
fishery is frozen for export to the US, where it is used in the production of fish feed
or pet food (Haig-Brown, 1994).
The proposed fisheries for M. norvegica, T. inermis and T. raschii off the East
coast of Canada are aimed at producing frozen krill. In addition, it is intended to
produce freeze-dried krill for ornamental fish and for public aquaria and freezedried krill as an ingredient in salmon feed and as a flavourant for food for human
consumption (Nicol & Endo, 1997). E. nana caught off the Uwajima Bay, South
East Japan, are used as feed for red sea bream (Y. Endo, pers. comm.).
Yields in the manufacture of products from Antarctic krill vary from nearly 100%
for krill hydrolysates, to 80±90% for fresh-frozen and boiled-frozen (Plate 6, facing
p. 182), to 8±17% for peeled krill and 10±15% for meal. There has been research
into ways of improving the efficiency of recovery in the Antarctic krill fishery and
reducing waste. Press waters and liquid by-products can contain significant amounts
266
Krill: Biology, Ecology and Fisheries
of protein and lipid, which can be removed by filtration, enzymatic or biotechnological methods (Dolganova, 1994).
Krill for human consumption
Individual krill products
Much of the available information on krill products for human consumption has
been summarised in earlier reviews (Budzinski et al., 1985; Suzuki & Shibata, 1990;
Plate 7, facing p. 182). Currently, 43% of the Japanese Antarctic krill catch is
processed for human consumption as boiled then frozen krill or peeled krill tail
meat frozen in blocks on board. Canned tail meat is no longer produced from the
Japanese catch. Information on products for human consumption from the Antarctic krill fisheries from nations other than Japan is not generally available. In the
past, Antarctic krill has also been used for the production of fermented protein
products, for spun protein products (i.e. surimi) (Suzuki & Shibata, 1990). A small
amount of E. pacifica caught off Japan is also being used for human consumption
(Kuroda, 1994).
Efforts have been made to produce low shell (hence fluoride) products. Krill
paste produced by traditional methods (Budzinski et al., 1985; Plate 8, facing p. 182)
and alkaline and acid processed krill protein concentrates have been produced in a
low fluoride form, by either organic acid washings or by simple water washings
(Tenuta-Filho, 1993). Using either treatment, fluoride concentrations of less than
21 mg g±1 (dry matter) were obtained, whereas untreated protein concentrates may
have values of ~ 250 mg g±1 (dry matter) (OehlenschlaÈger, 1981). These processes
yield high protein recovery and a product with low enough fluoride concentrations
for human consumption.
Considerable research has been carried out in Poland into producing krill precipitates using autoproteolysis, making use of krill's high level of proteolytic
enzymes to produce a high yield (80% protein recovery) concentrate (Kolakowski
& Gajowiecki, 1992). In this process, whole krill is mixed with water and heated.
The hydrolysate is centrifuged to remove the shells and the precipitate is coagulated. The final product has low fluoride content (< 29 mg F kg±1), a protein content
of 18±22%, fat less than 7% and a high level of carotenoid pigments, giving the
precipitate a pink-red colouration. This product is used mainly as a colorant and a
flavourant additive to fish feeds and other products for human consumption.
Krill as a food additive
Freeze-dried krill concentrate prepared from peeled tail meat is currently being
marketed as a food additive and as a health food supplement by a Spanish company.1 This `Antarctic Krill Concentrate' is advertised as having a number of useful
properties such as: high n-3 fatty acid content, moderate caloric content, high
Products Derived from Krill
267
nutritional value and ease of digestion, and is advertised as having a major revitalising function on the body. Suggested uses of this product as a dietary supplement
include during: pregnancy, lactation, pre- and post-menopausal stages, growth, postoperative procedures, cancer prevention, radiotherapy, chemotherapy, syndromes
of immunodeficiency and treatment of various nutritional disorders. Antarctic Krill
Concentrate is advertised by the manufacturer as containing important oligoelements, including antioxidants and minerals required to prevent dental cavities
and osteoporosis. The recommended dosage is approximately 5 g day±1. Antarctic
Krill Concentrate is promoted as being 100% natural and free of any side effects,
even when taken at higher dosages. The n-3 fatty acids in dehydrated krill products
are reported to remain unaltered even if stored for longer periods, retaining all their
beneficial properties.
Antarctic Krill Concentrate is produced as flakes, or as a loose powder, with
different degrees of granulation, and these have a light salmon-pink colour and an
excellent shrimp-like taste. It is promoted as an excellent natural colouring and
flavouring agent, which is effective even in small quantities when used in a variety of
foods. It is claimed to be suitable for the production of special dietary meals and
growth food products and requires no special storage conditions. Such specialty
products are likely to be of high value, but will utilise only small volumes of krill.
Aquaculture feed
Products
The development of krill products for human consumption has been a focus of the
Antarctic krill fishery in the past, but products for aquaculture are likely to dominate in the near future. Global aquaculture production more than doubled between
1986 and 1996 and currently accounts for over one-quarter of all fish consumed
(Naylor et al., 1998). Consequently, demand for quality aquaculture feed and feed
ingredients is growing rapidly and supplies are uncertain (Rumsey, 1993).
The existing or proposed coastal krill fisheries in the northern hemisphere have
been developed to provide local sources of feed for aquaculture and there have been
similar proposals in other areas, e.g. Nyctiphanes australis in South Eastern Australian waters (Virtue et al., 1995, 1996). It seems likely that stocks of other species
of krill will also be investigated once krill has become a more established aquaculture feed. Currently, most of the krill caught in all the commercial fisheries is
used for aquaculture feed. For Antarctic krill, 34% of the Japanese catch is fresh
frozen and 20% of this is used for aquaculture and 32% is used to produce meal
which is used in fish culture. Fifty per cent of the Japanese E. pacifica catch and
much of the Canadian catch of this species is used as an ingredient in feed for fish
culture.
One company, Specialty Marine Products, based in West Vancouver, Canada, is
planning a major project to harvest Antarctic krill for the burgeoning aquaculture
268
Krill: Biology, Ecology and Fisheries
industry. The company currently markets limited quantities of krill hydrolysates
(liquid and spray-dried) to the global livestock and aquaculture industries. The
company expects to begin its sustained harvest programme during 2000. The krill
will be processed on board the harvest vessels to ensure that the high standards for
product quality demanded by the global aquafeed industry will be met.
The use of euphausiids in aquafeeds has steadily increased in recent years (Nicol
& Endo, 1997; Storebakken, 1988). Aquafeed manufacturers have been including
krill products in feeds primarily to enhance the palatability of the feed, but krill is
also a good source of astaxanthin, a carotenoid pigment which is used to give a
characteristic red colour to some species.
Krill is available commercially in a variety of forms for use in aquafeeds: as a
meal, as frozen blocks of whole krill, and more recently as hydrolysates. Krill meal is
produced by much the same methods used in the manufacture of fish-meal
(Bykowski, 1986), whereas krill hydrolysates are prepared by the partial enzymatic
digestion of whole krill under controlled conditions. Freeze-dried krill and krill
hydrolysate have been used in fish feeds on an experimental basis, but their generally high costs have precluded large-scale use.
As with any ingredient, the freshness of the raw material and the type of processing influences the suitability of the final product for inclusion in feeds. This is
particularly the case for krill, since many of the components that contribute to the
flavour, as well as the pigments, are easily oxidised if exposed to excessive temperature during processing and drying. Typically, antioxidants are added prior to
processing to reduce this loss and to preserve the lipid quality (D. Saxby, pers.
comm.).
Krill species have a number of features that make them attractive ingredients for
aquaculture feeds (Anderson et al., 1997). These include: palatability enhancement
of feeds; a source of carotenoid pigments; a source of essential fatty acids; a wellbalanced amino acid profile; and an improvement of larval fish survival. To date, the
suitability of adding krill to feeds has been tested with a wide variety of fish species,
including: Atlantic and Pacific salmon (Salmo salar and Onchorynchus spp.); red sea
bream (Pagrus major); largemouth bass (Micropterus salmoides), eels (Anguilla
spp.); yellowtail (Seriola quinqueradiata); yellow perch (Perca flavescens); walleye
(Stizostedion vitreum); whitefish (Coregonus clupeaformis); sea bass (Dicentrarchus
labrax); sea bream (Sparus aurata) and Australian seabass (Lates calcarifer).
Adding krill hydrolysates to aquafeeds, even at levels of only a few per cent by
weight, make them more palatable (Forster, 1998), increasing feed consumption,
resulting in higher survival and weight gain. Using krill hydrolysate to improve
palatability enables the production of feeds that include higher levels of inexpensive
nutrient sources (e.g. plant and other protein sources) without affecting fish production (Oikawa & March, 1997).
There is considerable literature on the nutritional value of krill in cultured fish
diets (Storebakken, 1988). All species of krill that have been examined are highly
nutritious and can be used successfully as a source of protein, energy and flesh
Products Derived from Krill
269
pigmenting carotenoids for aquaculture species. Krill also contain a good balance of
amino acids and are effective feed stimulants as well. Krill-fed salmon were also
found to have a superior taste, but did not significantly accumulate fluoride from the
krill exoskeletons in their flesh. This paper (Storebakken, 1988), should be consulted for a review of the earlier literature.
Aquaculture feed additives
In recent years, several studies have been conducted to examine the value of krill
products in aquafeeds. The primary focus of this research has been to demonstrate
the effectiveness of krill as a feeding stimulant. Krill is known to have a positive
effect on the feeding behaviour of some fish. Shimizu et al. (1990) showed that diets
supplemented with krill meal stimulated feeding behaviour in sea bream (Pagrus
major) and that this effect was probably due to the presence of proline, glycine and
glucosamine. Not only does krill stimulate feeding, it promotes growth in some
species of fish (Allahpichay & Shimizu, 1985). The growth promoting factors seem
to be steroids located in the cephalothorax region, and thus are available in nonmuscle meal.
The use of krill (E. pacifica) as a food source for hatchery-reared salmon smolts
has contributed to increased disease resistance (Haig-Brown, 1994). This is attributed to the early development of the immune system when using krill as a food
source. The nutritive value of Nyctiphanes australis has recently been assessed with
regard to its possible use as an aquaculture feed by Virtue et al. (1995). These
researchers found that N. australis contained, on average, 52% protein and up to
9.5% lipid on a dry weight basis. The lipid content of N. australis was marked by the
presence of high quantities of unsaturated fatty acids with n-3 fatty acids accounting
for 49% of the total fatty acid content. Carotenoids were present at levels of up to
320 mg g±1 and were mainly (79.5%) in the form of astaxanthin. Fluoride levels were
as high as those reported in other species of krill (up to 3507 mg F g±1).
Krill hydrolysates are effective feeding stimulants in rainbow trout diets (Oikawa
& March, 1997). A control diet containing a mixture of fish-meal (20% of diet), and
plant protein sources (wheat (20%), soybean protein concentrate (10%), corn
gluten meal (10%), and canola meal (22%)) was compared to the same mixture
supplemented with 2% krill hydrolysate in replacement of fish-meal; and a third
diet, which also contained 2% krill hydrolysate, but this was blended into the oil and
coated on to the pellets. Krill hydrolysate significantly increased consumption of
diets in this trial and reduced feed wastage. Fish fed the second diet, with 2% krill
hydrolysate added, consumed 34% more feed and wasted less than half as much as
those fed the control feed containing no krill hydrolysate, as a proportion of total
feed input. This difference was even greater for fish fed the diet in which krill
hydrolysate was coated on to the pellets. In this case, the fish consumed 69% more
feed and again wasted less than half as much of the test feed, relative to those fed
control feed. This study further indicated that, because fish fed diets containing the
270
Krill: Biology, Ecology and Fisheries
krill hydrolysates consumed the feed more rapidly, they required less time to reach
satiation than those that were fed the control diet. The growth rates of the animals
followed the pattern of feed intake.
Krill products have been shown to be very effective in stimulating feeding in fish
that are in transition between life cycle stages. For example, weaning largemouth
bass (Micropterus salmoides) from live feeds to commercial dry pellets is facilitated
by feeding freeze-dried krill as a starter diet for four days (Kubitza & Lovshin,
1997). Following transfer to seawater, feed consumption by salmon smolts is often
inhibited, and it can take several weeks for full feeding to resume. Atlantic salmon
smolts fed a krill hydrolysate coated feed have higher feed intake, better feed
conversion efficiency (0.74 vs 0.87), faster growth, and lower percentage of `failed
smolt' than those fed a regular smolt feed (Santosh Lall of the Department of
Fisheries and Oceans Canada, pers. com.). The results of these trials were used in
the development of commercial smolt transfer diets.
Protein and lipid content
The protein content of krill is generally 60±65%, on a dry weight basis. The amino
acid profile of this protein is very well balanced with respect to fish-meal and the
requirements of cultured fish and crustacea (Table 10.1).
The digestibility of the protein of krill products is variable. The controlled
hydrolysis of many protein sources improves amino acid availability. A portion of
the amino acids in krill hydrolysates is present either in free form, or as short-chain
polypeptides, which tend to be more available than is intact protein. A comparison
of in vitro protein digestibility of krill hydrolysate and two krill meals, one from a
Japanese source and one from Russia, indicated considerable differences (I. Forster,
unpublished results). The Modified Torry Pepsin Digestibility coefficients of the
Japanese and Russian krill meals were 45 and 42%, respectively, while the digestibility coefficient of krill hydrolysate was 86%.
Lipid levels in krill products range from 10±20% on a dry weight basis. Krill oil is
rich in highly unsaturated fatty acids, most notably of the n-3 fatty acids (Table
10.2). These fatty acids are especially prone to degradation by oxidation and care
must be taken to ensure the freshness of the raw material and that appropriate
conditions are maintained during processing, including the addition of suitable
levels of antioxidants (Kolakowska, 1989, 1991b).
Pigments
Carotenoids levels in krill are around 30 mg g±1 and these appear to deteriorate
rapidly during storage if not refrigerated below 08C (Czerpak et al., 1980; Kolakowska, 1988). E. pacifica contains large amounts of carotenoid pigments, especially
astaxanthin. For this reason, up to 50% of the Japanese E. pacifica catch is used as
an ingredient in fish feeds to add a reddish colour to the skin and meat of fish species
32
3.8
1.3
2.3
3.1
4.5
2.0
4.4
1.8
0.4
2.6
38
3.9
1.8
2.4
3.7
4.7
2.6
4.7
2.1
0.5
3.2
Rainbow
trout1
38
5.4
1.6
2.0
3.5
4.5
3.6
4.6
2.0
0.4
2.9
Pacific
salmon1
35
3.7
1.8
2.2
2.9
5.0
2.7
5.7
3.4
0.7
3.1
Common
carp1
3
2
NRC, 1993
Guillaume, 1997
Rehbein, 1981
4
Biozyme Systems, Inc. internal data. Values are % of total amino acids of E. superba
5
Not determined
1
Total protein (% dry wt)
Arginine
Histidine
Isoleucine
Leucine
Lysine
Methionine + cysteine
Phenylalanine + tyrosine
Threonine
Tryptophan
Valine
Channel
catfish1
32
3.7
1.5
2.7
3.0
4.5
2.8
4.8
3.3
0.9
2.4
Tilapia1
40
5.8
2.1
3.4
5.4
5.3
3.6
7.1
3.6
0.8
4.0
Shrimp2
55
5.8
1.9
7.5
4.9
6.5
3.8
8.9
4.0
ND5
5.3
Krill
meal3
60
7.0
2.8
5.0
8.3
6.9
4.3
7.7
5.0
1.5
5.5
Krill
hydrolysate4
65.4
5.9
2.5
4.8
7.7
7.7
4.0
7.7
4.3
1.1
5.3
Anchovy
meal1
Table 10.1 Recommended protein and amino acid contents for cultured fish and shrimp compared to the protein and amino acid levels of krill meal,
krill hydrolysate and anchovy meal. Values are reported as percentage of protein.
Products Derived from Krill
271
272
Krill: Biology, Ecology and Fisheries
Table 10.2
superba)6.
Fatty acid
14 : 0
14 : 1n5
15 : 0
16 : 0
16 : 1n7
16 : 2
16 : 4n1
18 : 0
18 : 1n9
18 : 2n6
18 : 3n3
18 : 4n3
20 : 1n9
20 : 2n6
20 : 3n3
20 : 4n6
20 : 5n3
22 : 1n11
22 : 5n3
22 : 6n3
6
Fatty acid profile of krill hydrolysate (E.
Liquid
Spray-dried
5.6
0.1
0.3
18.0
4.2
0.4
0.5
0.8
15.1
2.0
1.0
1.4
0.4
0.3
0.2
0.7
18.1
1.4
0.4
14.7
5.5
0.0
0.0
21.6
5.3
0.0
0.0
2.7
22.2
4.8
1.2
1.5
1.0
0.0
0.0
1.2
18.1
0.0
0.4
11.9
Biozyme Systems, Inc. internal data
such as red sea bream, coho salmon, rainbow trout, yellowtail, and others (Kuroda,
1994). Extracts from Antarctic krill have also been successfully used as pigmenting
agents for yellowtail and coho salmon (Fujita et al., 1983a; Arai et al., 1987). As
Odate (1991) pointed out, red is traditionally viewed by the Japanese as an indication of good luck and red sea bream and lobsters are often used as offerings in
celebrations. Moreover, a reddish colour in fish meat is thought to stimulate the
appetite.
The principal pigment in krill is astaxanthin, although other carotenoid pigments
are also found (Czerpak et al., 1980; Czeczuga, 1981; Yamaguchi et al., 1983;
Kolakowska, 1988; Funk & Hobson, 1991). The level of astaxanthin can vary considerably among different krill products and species, but generally is between 150±
200 ppm (dry basis) for E. superba, while E. pacifica can contain considerably more.
Astaxanthin is present in free form, or esterified to either one or two fatty acids
(mono- and di-ester, respectively). By contrast, synthetic astaxanthin, which is
widely used in aquafeeds, is exclusively in the non-esterified form. The esterified
forms of astaxanthin must be converted to the free form prior to being absorbed
from the gut. Krill has been shown to be effective in pigmenting a variety of fish
(Fujita et al., 1983a, b; Arai et al., 1987), although this has not always been found to
be the case in salmonids (Foss et al., 1987; Whyte et al., 1998). Recent work with
Atlantic salmon compared the bioavailability of astaxanthin from spray-dried krill
Products Derived from Krill
273
hydrolysate, krill meal and synthetic astaxanthin at levels in the experimental diets
up to 50 ppm. The astaxanthin from krill hydrolysate was 91% available as the
synthetic source (I. Forster, unpublished data), whereas the bioavailability of krill
meal was much less. Astaxanthin is very susceptible to degradation from high heat,
such as occurs during the manufacture of aquafeeds. In general, under these conditions the destruction is expected to be about 10±15%, and is about the same for
free and esterified forms.
Minerals
Krill products are a good source of minerals for aquatic animals (Table 10.3).
Recent work has shown that rainbow trout fed feeds containing krill as the principal
protein source had significantly less dorsal fin erosion than did those fed the fishmeal based control feed (Lellis & Barrows, 1997). Fin erosion is a common problem
among cultured salmonids, and may negatively affect survival, disease resistance,
desirability among sportfish anglers and the commercial value of the fish. This
beneficial effect of krill is thought to be derived from the balance and availability of
micro minerals (Rick Barrows, US Dept. of Fish and Wildlife, Bozeman, Montana,
pers. comm.).
Table 10.3 The mineral content of krill hydrolysate
(E. superba).
Mineral
Level7
Sodium
Calcium
Phosphorus
Potassium
Sulfur
Magnesium
Zinc
Iron
Copper
2.6%
2.3%
1.9%
1.3%
1.3%
0.4%
40 ppm
36 ppm
33 ppm
7
Biozyme Systems, Inc. internal data. Values are on a dry
weight basis.
Krill as shrimp feed
Little work has been done to date, to investigate the potential of krill products in
feeds for shrimp. Krill products also provide many of the nutrients known to be
important for shrimp; e.g. amino acid, fatty acid and minerals. In addition, they are a
good source of astaxanthin, and it is believed that they possess excellent olfactory
attractant characteristics, which is of particular relevance to companies involved in
the development of penaeid shrimp feeds. Much work remains to be done in this
area.
274
Krill: Biology, Ecology and Fisheries
Uses of krill in aquafeeds
Krill products are added to feeds for aquatic animals in a variety of company and
product specific ways. Some feed manufacturers add whole krill to the feed mixture
prior to extrusion, in partial replacement of water. Other companies add dried krill
by spraying a blend of the powder and fish oil on to the pelleted feed. Krill products
(freeze-dried krill and krill hydrolysate) have been added directly to culture water
to stimulate feeding of larval fish.
A new product which uses the enzymes of krill to hydrolyse and liquefy krill,
before pelletising and drying is being developed in British Columbia for salmon
farming and may be applicable to krill fisheries elsewhere (Haig-Brown, 1994). A
Canadian firm, Biozyme Systems, Inc., is producing these high value krill hydrolysate products using a proprietary process it has developed. These products are
offered in liquid, concentrated, dried or frozen forms. The markets for these
products are in the animal and aquaculture feed industries (D. Saxby, pers.
comm.2).
Sport fishing
A considerable percentage of the Japanese Antarctic krill catch is used for sport
fishing bait. Fresh frozen krill comprises 34% of the total catch and of this 70% is
sold whole as bait and 10% of this is used as chum for sport fishing. Kuroda (1994)
suggests that there is little competition between E. superba and E. pacifica used for
sport fishing, because smaller E. pacifica is used as a chum (about 50% of total
catch), whereas the larger E. superba is mostly used as a bait.
Aquarium feed
A small quantity of Antarctic krill is freeze-dried for the home aquarium market.
An estimated 50% of the catch of E. pacifica from the British Columbia fishery ends
up in aquarium food (Dave Barrett, pers comm.3). Krill for ornamental fish feed are
marketed on the basis of their high quality proteins and n-3 fatty acids4. Fish feeds
using euphausiids are promoted as a protein source that is digestible, increases the
colour, improves disease resistance, stimulates breeding and enhances growth of
fish. Krill can be combined with other high biological-value nutrients (e.g. Spirulina
algae), as a basis for top quality fish feeds.
Euphausiids for aquarium use are available in several forms: freeze-dried, frozen
or as an ingredient in flake and pelleted feeds and are marketed as a feed to be used
when the fish have not been eating, due to being moved or when recovering from a
disease problem.
Products Derived from Krill
275
Chemicals and pharmaceuticals
Introduction
There has been a steady development of krill products for non-nutritional uses such
as for pharmaceuticals and for industrial uses (Nicol & Endo, 1997). These include
the production of chitin and chitosan from krill shells and the utilisation of krill
enzymes for pharmaceutical (Melrose et al., 1995; Mekkes et al., 1997) and other
purposes (Makes, 1992). These demands may not develop to the point where they
become a major economic justification for krill fishing, but they may be instrumental in putting the fishery on a sound economic footing.
Chitin
Krill is a source of chitin (Anderson et al., 1997) and considerable research has been
carried out into the extraction of this chemical from Antarctic krill (Breski, 1989).
The chitin content of krill and the annual production of chitin by krill has been
reviewed by Nicol & Hosie (1993) and the chitin composition of whole Antarctic
krill is between 2.4 and 2.7% of the dry weight.
Chitin and chitosan, which is derived from chitin, have a wide variety of current
and potential uses, ranging from loudspeaker membranes to cholesterol lowering
applications (Sandford, 1989; Nicol, 1991; Maezaki et al., 1993; Peter, 1995), which
might become lucrative by-products of the krill fishing industry. It is unlikely,
however, that a krill fishery would be initiated solely to produce chitin.
Lipids
The lipid composition of Antarctic krill has been reviewed by Suzuki & Shibata
(1990). More detailed analyses of lipids have been carried out recently. Antarctic
krill caught in winter were analysed for their fatty acid composition by Kolakowska
et al. (1994). Polyunsaturated n-3 fatty acids were found to comprise 19% of the
total fatty acids and were stable during processing, making krill an attractive
nutritional source of these fatty acids. Winter-caught krill had similar levels of fatty
acids to their summer-caught counterparts but there was less variability, probably
because of the lack of reproductive activity which can raise the lipid content of
mature females to more than 8% of their wet weight (Kolakowska, 1991a). Winter
season krill contained 3% wet weight lipids (Kolakowska et al., 1994) and the highly
unsaturated fatty acids, eicosapentaenoic acid (EPA) and dodecahexaenoic acid
(DHA), accounted for about 19% of the total fatty acids. Kolakowska (1991b)
concluded that E. superba, and even the waste products from the current Polish
processing technology, is a valuable source of n-3 polyunsaturated fatty acids. The
lipids are more stable, and contain much higher levels of carotenoid pigments than
some fish-meals.
276
Krill: Biology, Ecology and Fisheries
The chemical composition of E. pacifica has not been studied as intensively as E.
superba. The fatty acid composition of E. pacifica is similar to that of E. superba,
with 14:0 and 16:0 being the main saturated acids (10±20% of the total fatty acids),
whereas 16:1 and 18:1 are the main monounsaturated fatty acid components (6%
and 15%, respectively). E. pacifica contains higher amounts of EPA (c. 23%) and
DHA (c. 14%) than does E. superba (Yamada, 1964).
The lipid composition of the three major north Atlantic species of krill, M. norvegica, T. raschii and T. inermis, have been examined in detail (Saether et al., 1986).
The fatty acid composition appears to be related to the diet, which varies seasonally,
but these three species contain more depot lipids ± triacylglycerides in M. norvegica,
and wax esters and glycerophospholipids in the two Thysanoessa species ± than does
Antarctic krill.
N. australis from coastal Tasmania has lower lipid levels than do higher latitude
species (8.5% dry weight vs 15±50%) and this level varies little with season (Virtue
et al., 1995). Such lower level of total lipid may have advantages when it comes to
using this species as an aquaculture feed. In terms of lipid composition, N. australis
contains very high levels of long chain n-3 polyunsaturated fatty acids, which are
essential for healthy growth in salmonids, and has higher levels of the carotenoid
pigment astaxanthin than does E. superba. Both these features favour its use as an
aquafeed ingredient despite its lower protein value (52% dry weight) than traditional salmon feed meals (60±70%).
Enzymes
Krill contain very effective hydrolytic enzymes, including proteases, carbohydrases,
nucleases and phospholipases. These enzymes appear to be concentrated in the
digestive gland in the cephalothorax of the krill. These enzymes have found medical
uses in debriding necrotic tissue (Anheller et al., 1989) and as chemonucleolytic
agents (Melrose et al., 1995).
Three approaches have been taken to producing an enzymatic debriding agent for
the treatment of necrotic wounds (Karlstam et al., 1991). In the first process, whole
krill are de-oiled and the proteolytic enzymes are isolated and purified by sizeexclusion chromatography. The second process also de-oils the raw krill, followed
by homogenisation of the residue, further de-oiling of the aqueous solution and
precipitation using an organic solvent. This results in an enzyme powder with low
specific activity. In the third process, the enzymes are purified from squeezed
autolysed krill. The krill extract can be used directly, or after further purification of
individual enzymes.
Krill digestive proteases have also been examined as potential chemonucleolytic
agents (Melrose et al., 1995) and proteases from Antarctic krill have shown considerable potential. Chemonucleolysis is a therapeutic procedure whereby a
degradative agent is injected to reduce the height of vertebral discs and diminish
disc pressure on inflamed nerve roots in cases of sciatica.
Products Derived from Krill
277
In another development of the medical application of enzymes from Antarctic
krill, Phairson Medical has identified and purified a single enzyme from krill
(PHM-101), and is moving through the lengthy steps of drug development with
this compound. They have identified the molecular mechanism of action of this
enzyme. It acts on a wide range of protein cell adhesion molecules and several
other clinically relevant molecules, and it appears to readily cut CD4, CD8,
ICAM 1 & 2, E & L Selectin and VCAM, all of which are important in inflammation. It also cuts an adhesion molecule on Candida albicans, as well as
degrades IL-2, bradykinin, and staphylococcus enterotoxin B. This broad specificity may result from the intrinsic flexibility of the molecule. It is a serine protease, with properties common to chymotrypsin, trypsin, collagenase and elastase.
It functions over a wide pH range and is relatively stable up to 558C. Clinically, it
has shown a response in the treatment of oral and vaginal candida infections,
acne and wound care and may have even broader uses. (R. Franklin, Phairson
Medical, pers. comm.5).
Krill enzymes have also found a use as agents for assisting the restoration of works
of art (Makes 1992). Development of markets for high value products, such as krill
enzymes, along with the production of food items or aquaculture feed ingredients,
will stimulate the growth of krill fisheries.
Economics
Development of products from krill fisheries is highly cost-dependent, but the
economics of krill fisheries has not been subject to close scrutiny recently. The
market value of E. superba is difficult to ascertain accurately, but whole frozen krill,
in 1996 prices, fetched approximately Aus.$0.32 kg±1, whereas frozen tail meat was
reported to fetch around Aus.$9.50 kg±1 (figures from industry sources). No published economic analysis on the Antarctic krill fishery has been attempted since the
review in Budzinski et al. (1985). The low level of fishing for Antarctic krill in the
last five years has been attributed to the high costs and low returns associated with
standard catching and processing technology.
The value of euphausiids landed in British Columbia varied between Can$0.23
and $0.88 kg±1 between 1984 and 1994 and was Can$0.55 ± $0.88 kg±1 in 1995. The
total landed value of the euphausiid fishery in British Columbia is comparatively
small, varying between Can$28 000 in 1985 and $415 000 in 1990. The 1995 catch was
worth approximately Can$357 000.
The average price of E. pacifica on landing in Japan for 1989±1993 was ¥2.7 billion
in total. The average unit price was ¥45.4 kg±1, with a lower price of ¥21±36 kg±1 in
the last 2 years, because of the very high catch in 1992 (Kuroda & Kotani, 1994). The
availability of Antarctic krill does not seem to affect the price of E. pacifica, as long
as the production level of the former species does not change drastically as shown by
multiple regression analysis (Yoshida, 1995). Yoshida (1995) showed that the
average unit price of E. pacifica (Y in ¥ kg±1) can be predicted by the cumulative
278
Krill: Biology, Ecology and Fisheries
catch at the end of March (X1 in kT) and the annual catch of the previous year (X2 in
kT) as follows:
Y = 71.15X1 7 0.35X2 + 99.41
with multiple correlation coefficient of 0.89. This means that the increase in X1 by
10 kT causes prices to be reduced by 11.5 ¥ kg±1, and the same increase in X2 causes a
price reduction of 3.5 ¥ kg±1. There is a good agreement between the real average
price and the calculated value during the years 1987±1994 (Nicol & Endo, 1997).
10.4
Trends and future developments
Considerable innovation has been displayed to develop products from krill that will
provide a reasonable economic return. Harvesting and processing technology has
improved markedly over the history of the fishery, but the costs of embarking on an
Antarctic fishery are still largely prohibitive. The historic concentration of this
fishery on producing products for human consumption may change as innovative
aquafeed products, such as hydrolysates, gain acceptance and demand for aquafeeds
grows. In 1993 12% of the world's fish-meal was used for aquaculture feeds and a
projection to the year 2000 saw the requirement rise to 10±25% of fish-meal production (Rumsey, 1993). In 1997, about 1.8 million tonnes of wild fish were required
to produce the fish-meal and oil used to produce 644 kT of Atlantic salmon (Naylor
et al., 1998). Other farmed species have similar conversion ratios. Between 1970 and
1990 world production of fish-meal had increased 27%, and was projected to decline
by 5% between 1990 and 2000, despite an annual growth rate of aquaculture of
about 12%. The development of aquafeeds using sources of protein other than fishmeal has received considerable attention in recent years. The growth rate of
aquaculture, and the trend towards intensive rather than subsistence production
methods, will put greater demands on existing fisheries for fish-meal ± many of
which are currently over-exploited ± and may drive a move toward exploiting
alternate sources of aquafeed ingredients, such as krill. All krill species have a
number of features that make them attractive as source materials for the aquaculture feed industry. As these features become more well-known, the demand for
krill will grow. The prices of krill will tend to decrease as increases in the size of krill
fisheries improve the economies of scale. Although much of the development of
new aquafeed products is occurring in the Northern Hemisphere (Haig-Brown,
1994), management caution in the development of so-called `forage fisheries' is
tending to prevent their development in coastal and shelf areas. Thus, if demand is
generated for krill products for aquaculture, this will only be able to be met from the
Antarctic region.
It is likely that krill products for human consumption will continue to be produced, but they will not be the driving force behind the fishery in the near future. In
Products Derived from Krill
279
parallel with the development of krill products for aquaculture and human consumption, there will be a likely increase in demand for krill products for nonnutritional uses, such as for pharmaceuticals and for industrial uses. Some of these
demands may develop to the point where they become a major economic justification for krill fishing, or at least will be instrumental in placing krill fisheries on a
sound economic basis.
Notes
1
Information provided by: C. Falkenberg, Monteclaro Asesores, S.L. Acacias 10, Monteclaro, 28223 Madrid, Spain.
2
D. Saxby, Biozyme Systems, Inc., West Vancouver, BC V7V 1N6 Canada.
3
D. Barratt, Murex Aqua Foods, Inc.
4
Aquatrol inc. http://www.aquatrol.com/aboutus.html
5
Phairson Medical Ltd., 602 The Chambers, Chelsea Harbour, London, SW10 0XF, UK.
References
Adelung, D., Buchholz, F., Culik, B. & Keck, A. (1987) Fluoride in tissues of krill Euphausia
superba Dana and Meganyctiphanes norvegica M. Sars in relation to the moult cycle. Pol.
Biol. 7, 43±50.
Allahpichay, I. & Shimizu, C. (1985) Separation of growth promoting factors from nonmuscle krill meal of Euphausia superba. Bull. Japanese Soc. Sci. Fish. 51, 945±51.
Anderson, S., Richardson, N.L., Higgs, D.A. & Dosanjh, B.S. (1997) The evaluation of airdried whole krill meal as a dietary protein supplement for juvenile Chinook salmon
(Oncorhynchus Tshawytscha). Canadian Technical Report of Fisheries and Aquatic
Sciences 2148.
Anheller, J.-E., Hellgren, L., Karlstam, B. & Vincent, J. (1989) Biochemical and biological
profile of a new enzyme preparation from Antarctic krill Euphausia superba Dana suitable
for debridement of ulcerative lesions. Arch. Dermatol. Res. 281, 105±110.
Arai, S., Mori, T., Miki, W., Yamaguchi, K., Konsu, S., Satake, M. & Fujita, T. (1987) Pigmentation of Juvenile Coho salmon with carotenoid oil extracted from Antarctic krill.
Aquaculture 66, 255±64.
Breski, M.M. (1989) Production and application of chitin and chitosan in Poland. In Chitin
and Chitosan. Sources, chemistry, biochemistry, physical properties and applications (G.
Skjak-Braek, T. Anthonsen & P. Sandford, eds). Elsevier Science, New York, USA.
Budzinski, E., Bykowski, P. & Dutkiewicz, D. (1985) Possibilities of processing and marketing of products made from Antarctic krill. FAO Fisheries Technical Paper 268. pp. 1±46.
Bykowski, P.J. (1986) Possibilities of utilising Antarctic krill. Infofish. Marketing Digest 6/86,
11±13.
CCAMLR (1993) Scientific Observer's Manual (Pilot Edition). Commission for the Conservation of Marine Living Resources, Hobart, Australia, 68 pp.
Christians, O. & Leinemann, M. (1983) Investigations on the migration fluoride from the shell
into the muscle flesh of Antarctic krill (Euphausia superba) in dependence of storage
temperature and time. Archiv fur Fischereiwisschaffen 34, 87±95.
280
Krill: Biology, Ecology and Fisheries
Czeczuga, B. (1981) The presence of carotenoids in some invertebrates of the Antarctic coast.
Comp. Biochem. Physiol. 69B, 611±15.
Czerpak, R., Jackowska, H. & Mical, A. (1980) Qualitative analysis of carotenoids in particular parts of body of males and females of Euphausia superba Dana (Crustacea). Polish
Polar Res. 1, 139±45.
Dolganova, N. (1994) Comprehensive processing of krill and fish press waters and other
liquid by-products. Acta Ichthyologica et Piscatoria XXIV, 171±77.
Donachie, S.P. & Zdanowski, M.K. (1998) Potential digestive function of bacteria in krill
Euphausia superba stomach. Aquat. Micro. Ecol. 14, 129±36.
Donachie, S.P., Sabrowski, R., Peters, G. & Buchholz, F. (1995) Bacterial enzyme activity in
the stomach and hepatopancreas of Meganyctiphanes norvegica (M. Sars, 1857). J. Exp.
Biol. Ecol. 188, 151±65.
Eddie, G.C. (1977) The harvesting of krill. Southern Ocean Fisheries Survey Programme.
FAO Rome. GLO/SO/77/2, 1±76.
Everson, I. (1977) The living resources of the Southern Ocean. Southern Ocean Fisheries
Survey Programme. FAO Rome. GLO/SO/77/1, 1±156.
Fevolden, S.E. (1981) Bacteriological characteristics of Antarctic krill (crustacea, euphausiacea). Sarsia 66, 77±82.
Fisher, L.R., Kon, S.K. & Thompson, S.Y. (1953) Vitamin A and carotenoids in some
Mediterranean crustacea with a note on the swarming of Meganyctiphanes. Bull. Inst.
Oceanogr. Monaco 1012, 1±19.
Forster, I. (1998) Krill hydrolysates. Int. Aquafeeds 4, 21±24.
Foss, P., Storebakken, T., Austreng, E. & Liaaen-Jensen, S. (1987) Carotenoids in diets for
salmonids. V. Pigmentation of rainbow trout and sea trout with astaxanthin and astaxanthin dipalmitate in comparison with canthaxanthin. Aquaculture 65, 293±305.
Fujita, T., Satake, M., Hikichi, S. et al. (1983a) Pigmentation of cultured yellowtail with krill
oil. Bull. Japanese Soc. Sci. Fish. 49, 1595±1600.
Fujita, T., Satake, M., Watanabe, T. et al. (1983b) Pigmentation of cultured red sea bream
with astaxanthin diester purified with krill oil. Bull. Japan. Soc. Sci. Fish. 49, 1855±61.
Funk, V.A. & Hobson, L.A. (1991) Temporal variations in the carotenoid composition and
content of Euphausia pacifica Hansen in Saanich Inlet, British Columbia. J. Exp. Mar. Biol.
Ecol. 148, 93±104.
Gajowiecki, L. (1995) Effects of feeding intensity of Antarctic krill (Euphausia superba
Dana) on qualitative and quantitative composition of its volatile compounds. Adv. Agr. Sci.
4, 75±82.
Grantham, G.J. (1977) The Southern Ocean: The Utilization of Krill. Southern Ocean Fisheries Survey Programme. FAO Rome. GLO/SO/77/3, 1±61.
Grave, H. (1981) Fluoride content of salmonids fed on Antarctic krill. Aquaculture 24, 191±96.
Haig-Brown, A. (1994) Going for the krill. National Fisherman, May 1994, 18±19.
Hays, R., Measures, L.N. & Huot, J. (1998) Euphausiids as intermediate hosts of Aniaskis
simplex in the St. Lawrence estuary. Can. J. Zool. 76, 1226±35.
Kagei, N. (1985) Krills and their parasites (abstract only). Bull. Mar. Sci. 37, 768.
Kagei, N., Asano, K. & Kihata, M. (1978) On the examination against the parasites of
Antarctic krill Euphausia superba. Sci. Repts. Whale Res. Inst. Tokyo 30, 311±13.
Karlstam, B., Vincent, J. Johansson, B. & Bryno, C. (1991) A simple purification method of
squeezed krill for obtaining high levels of hydrolytic enzymes. Prep. Biochem. 21, 237±56.
Products Derived from Krill
281
Kolakowska, A. (1988) Changes in lipids during the storage of krill (Euphausia superba
Dana) at 38C. Z. Lebensm. Unters. Forsch. 186, 519±23.
Kolakowska, A. (1989) Krill lipids after frozen storage of about 1 year in relation to storage
time before freezing. Die Nahrung 33, 241±44.
Kolakowska, A. (1991a) The influence of sex and maturity stage of krill (Euphausia superba
Dana) upon the content and composition of its lipids. Polish Pol. Res. 12, 73±78.
Kolakowska, A. (1991b) The oxidability of krill lipids. Polish Pol. Res. 12, 89±104.
Kolakowska, A., Kolakowski, E. & Szczygielski, M. (1994) Winter season krill (Euphausia
superba Dana) as a source of n-3 polyunsaturated fatty acids. Die Nahrung 38, 128±34.
Kolakowski, E. (1989) Comparison of krill and Antarctic fish with regard to protein solubility.
Z. Lebensm. Unters. Forsch. 188, 419±25.
Kolakowski, E. & Gajowiecki, L. (1992) Optimization of autoproteolysis to obtain an edible
product `precipitate' from Antarctic krill (Euphausia superba Dana). In Seafood Science
and Technology (E.G. Bligh, ed.). Fishing News Books, Oxford.
Kotarbinska, M. & Grosyzk, K. (1977) Evaluation of krill meal as a feed for growing pigs.
Nowe rolnice 26, 27±29. (In Polish.)
Krasowska, A. (1989) Influence of low-chitin krill meal on reproduction of Clethrionomys
glareolus (Schreber, 1780). Comp. Biochem. Physiol. 94C(1), 313±20.
Kubitza, F. & Lovshin, L.L. (1997) Effects of initial weight and genetic strain on feed training
largemouth bass Micropterus salmoides using ground fish flesh and freeze dried krill as
starter diets. Aquaculture 148, 179±90.
Kulka, D.W. & Corey, S. (1984) Incidence of parasitism and irregular development of gonads in
Thysanoessa inermis (Kroyer) in the Bay of Fundy (Euphausiacea). Crustaceana 46, 87±94.
Kuroda, K. (1994) Euphausiid fishery in the Japanese waters and related research activities.
Gekkan Kaiyo 26, 210±217.
Kuroda, K. & Kotani, Y. (1994) Euphausiid fishery and interannual variation of the fishing
conditions in the Sanriku-Joban coastal areas. Gekkan Kaiyo 26, 210±217.
Lellis, W.A. & Barrows, F.T. (1997) The effect of diet on dorsal fin erosion in steelhead trout
(Oncorhynchus mykiss). Aquaculture 156, 229±40.
Lotz, J.M. (1997) Viruses, biosecurity and specific pathogen-free stocks in shrimp. World J.
Microbiol. Biotech. 13, 405±413.
Maezaki, Y., Tsuji, K., Nakagawa, Y. et al. (1993) Hypocholesterolemic effect of chitosan in
adult males. Biosci. Biotech. Biochem. 57, 1439±44.
Makes, F. (1992) Enzymatic examination of the authenticity of a painting attributed to
Rembrant. Krill enzymes as diagnostic tool for identification of `The Repentant
Magdlene', Institute of Conservation of the University of Gothenburg, Gothenburg.
Mauchline, J. & Fisher, L.R. (1969) The biology of euphausiids. Adv. Mar. Biol. 7, 1±454.
Mekkes, J.R., Le Poole, I.C., Das, P.K., Kammeyer, A. & Westerhof, W. (1997) In vitro
tissue-digesting properties of krill enzymes compared to fibrinolysin/DNAs, papaine and
placebo. Int. J. Biochem. Cell Biol. 29, 703±706.
Melrose, J., Hall, A., Macpherson, C., Bellenger, C.R. & Ghosh, P. (1995) Evaluation of
digestive proteinases from the Antarctic krill Euphausia superba as potential chemonucleolytic agents. In vitro and in vivo studies. Arch. Orthop. Trauma Surg. 114, 145±52.
Miller, D.G.M. (1991) Exploitation of Antarctic marine living resources: A brief history and a
possible approach to managing the krill fishery. S. Afr. J. Mar. Sci. 10, 321±39.
Naylor, R.L., Goldburg, R.J., Mooney, H.M., Beveridge, M., Clay, J., Folke, C., Kautsky, N.,
282
Krill: Biology, Ecology and Fisheries
Lubchenko, J., Primavera, J. & Williams, M. (1998) Nature's subsidies to shrimp and salmon farming. Science 282, 883±84.
Neal, R.A. & Maris, R.C. (1985) In The biology of crustacea. Fisheries biology of shrimps and
shrimplike organisms (A.J. Provenzano, ed.). Academic Press, London.
Nemoto, T. (1970) On the parasitic organisms in a krill, Euphausia similis, from Suruga Bay.
J. Oceanogr. Soc. Japan 26, 283±95.
Nicol, S. (1984) Ephelota sp. a Suctorian found on the euphausiid Meganyciphanes norvegica.
Can. J. Zool. 62, 744±46.
Nicol, S. (1991) Life after death for empty shells. New Scientist 1755, 36±38.
Nicol, S. & Endo, Y. (1997) Krill Fisheries of the World. FAO, Rome, 110 pp.
Nicol, S. & Endo, Y. (1999) Krill fisheries: development, management and ecosystem
implications. Aquat. Liv. Resour. 12, 1±17.
Nicol, S. & Hosie, G.W. (1993) Chitin production by krill. Biochem. System. Ecol. 21, 181±84.
NRC (1993) National Research Council. Nutrient Requirements of Fish. National Academy
Press, Washington, D.C., 114 pp.
Odate, K. (1991) Fishery biology of the krill, Euphausia pacifica, in the Northeastern coasts of
Japan. Suisan Kenkyu Sosho 40, 1±100.
OehlenschlaÈger, J. (1979) First attempt to feed krill to mink. Informationen fur die Fischwirtschaft 26, 48. (In German.)
OehlenschlaÈger, J. & Manthey, M. (1982) Fluoride content of Antarctic marine animals
caught off Elephant Island. Polar Biol. 1, 125±27.
Oehlenschlager, J.a.S. (1981) A functional protein concentrate (KPC) from Antarctic krill
(Euphausia superba, Dana 1850). Preparation, chemical composition and functional
properties. Z. Lebensm. Unters. Forsch. 172, 393±98.
Oikawa, C.K. & March, B.E. (1997) A method for assessment of the efficacy of feed
attractants for fish. Prog. Fish Culturist 59, 213±17.
Omori, M. (1978) Zooplankton fisheries of the world: A review. Mar. Biol. 48, 199±205.
Parsons, T.R. (1972) Plankton as a food source. Underwater J. 4, 30±37.
Peter, M. (1995) Applications and environmental aspects of chitin and chitosan. J.M.S. ± Pure
Appl. Chem. A32, 629±40.
Peters, G., Saborowski, R., Mentlein, R. & Buchholz, F. (1998) Isoforms of an n-acetyl-betaD-glucosaminidase from the Antarctic krill, Euphausia superba: purification and antibody
production. Comp. Biochem. Physiol. B-Biochem. Mol. Biol. 120, 743±51.
Rakusa-Suszczewski, S. & Filcek, K. (1988) Protozoa on the body of Euphausia superba Dana
from Admiralty Bay (the South Shetland Islands). Acta Protozoologica 27, 21±30.
Rakusa-Suszczewski, S. & Zadonowski, M. (1980) Decomposition of Euphausia superba
Dana. Pol. Arch. hydrobiol. 27, 305±311.
Rehbein, H. (1981) Amino acid composition and pepsin digestibility of krill meal. J. Agric.
Food Chem. 29, 682±84.
Rumsey, G.L. (1993) Fish meal and alternate sources of protein in fish feeds. Update 1993.
Fisheries 18, 14±19.
Saether, O., Ellingsen, T.E. & Mohr, V. (1986) Lipids of North Atlantic krill. J. Lipid Res. 27,
274±85.
Sandford, P. (1989) Chitosan: commercial uses and potential applications. In Chitin and
Chitosan. Sources, chemistry, biochemistry, physical properties and applications (G. SkjakBraek, T. Anthonsen & P. Sandford, eds). Elsevier Science, New York.
Products Derived from Krill
283
Sands, M., Nicol, S. & McMinn, A. (1998) Fluoride in Antarctic marine crustaceans. Mar.
Biol. 132, 591±98.
Schneppenheim, R. (1980) Concentration of fluoride in Antarctic animals. Meersforchung.
Reports on Marine Research 28, 179±82.
Shimizu, C., Ibrahim, A., Toroko, T. & Shirakawa, Y. (1990) Feeding stimulation in sea
bream, Pagrus major, fed diets supplemented with Antarctic krill meals. Aquaculture 89,
43±53.
Ê ., Karlstam, B., Vincent, J. & Roeraade, J. (1998) Separation of
Sjodahl, J., Emmer, A
proteolytic enzymes originating from Antarctic krill (Euphausia superba) by capillary
electrophoresis. J. Chromatography B 705, 231±41.
Soevik, T. & Breakkan, O.R. (1979) Fluoride in Antarctic krill (Euphausia superba) and
Atlantic krill (Meganyctiphanes norvegica). J. Fish. Res. Bd. Can. 36, 1414±16.
Storebakken, T. (1988) Krill as a potential feed source for salmonids. Aquaculture 70, 193±
205.
Suzuki, T. (1981) Fish and Krill Protein: Processing Technology. Allied Science Publishers,
pp. 193±251.
Suzuki, T. & Shibata, N. (1990) The utilization of Antarctic krill for human food. Food
Reviews International 6, 119±47.
Tenuta-Filho, A. (1993) Fluorine removal during production of krill paste and krill protein
concentrates. Acta Alimaentaria 22, 269±81.
Virtue, P., Johannes, R.E., Nichols, P.D. & Young, J.W. (1995) Biochemical composition of
Nyctiphanes australis and its possible use as an aquaculture feed source: lipids, pigments
and fluoride content. Mar. Biol. 122, 121±28.
Virtue, P., Johannes, R.E., Nichols, P.D. & Young, J.W. (1996) The biochemical composition
of Nyctiphanes australis and its possible use as an aquaculture feed source: lipids, pigments
and fluoride content pp. 1±4. CSIRO Divisions of Oceanography and Fisheries, Marine
Laboratories, Hobart, Tasmania, Australia.
Virtue, P., Nichols, P.D. & Nicol, S. (1997) Dietary related mechanisms of survival in
Euphausia superba: changes during long term starvation and bacteria as a possible source
of nutrition. In Antarctic Communities ± species, structure and survival (B. Battaglia, J.
Valencia & D.W.H. Walton, eds). Cambridge University Press, Cambridge.
Whyte, J.N.C., Travers, D. & Sherry, K.L. (1998) Deposition of astaxanthin isomers in chinook salmon (Oncorhynchus tshawytscha) fed different sources of pigment. Fisheries and
Oceans, Canada, Nanaimo, B.C.
Yamada, M. (1964) On the lipids of plankton. Bull. Jap. Soc. Sci. Fish. 30, 673±81.
Yamaguchi, K., Miki, W., Toriu, N., Kondon, Y., Murakami, M., Konosu, S., Satake, M. &
Fujita, T. (1983) The composition of carotenoid pigments in the Antarctic krill Euphausia
superba. Bull. Japan. Soc. Sci. Fish. 49, 1411±15.
Yoshida, H. (1995) A study on the price formation mechanism of Euphausia pacifica. Bull.
Jap. Soc. Fish. Oceanogr. 59, 36±38.
Chapter 11
Management of Krill Fisheries in
Japanese Waters
Yoshi Endo
Three euphausiid species are commercially exploited in the Japanese waters:
Euphausia pacifica Hansen, E. nana Brinton and Thysanoessa inermis (Kroyer)
(Fig. 11.1). The most important is E. pacifica or, in the Japanese vernacular,
`tsunonashi-okiami' (krill without projection on the head), with an average annual
catch of 62 192 t having been taken for the past 10 years. The total landed value of
Fig. 11.1
284
Euphausiid fisheries in Japanese waters.
Krill Fisheries in Japanese Waters
285
E. pacifica is 1.5±4.4 billion yen per year and the fishery is now very important in the
Japanese coastal fisheries. The E. pacifica fishery is the second largest krill fishery in
the world following E. superba fishery in the Antarctic Ocean (Nicol & Endo, 1997).
In Iwate Prefecture, annual catch of E. pacifica ranked second of all the sea
fisheries following salmonid fish in 1994, 1995 and 1997, and ranked third following
salmonid fish and squids in 1996; the landed values of E. pacifica ranked in the top
seven following salmonid fish, tunas, squids, abalones and so on (Iwate Prefecture,
1999). In Miyagi Prefecture, annual catch of E. pacifica ranked 4th±6th following
tunas plus marlins, squids, bonito and so on during the period 1993±1996; the landed
value ranked in the top nine following tunas plus marlins, squids, bonito, gadoid fish,
flatfish, salmonid fish and so on (Miyagi Prefecture, 1998). In the southern prefectures, only small amounts of E. pacifica have been landed during these years
because of warmer oceanographic conditions, as will be discussed in Section 11.3.
11.1
History of the Euphausia pacifica fishery
The Euphausia pacifica fishery developed during the Meiji Era, approximately the
last 100 years, from a fishery for sand lance (Ammodytes personatus) using a bowmounted trawl in Sanriku coastal waters (Odate, 1991). The same method was used
to harvest E. pacifica in the mid-1940s by the fishermen of the Oshika Peninsula in
Miyagi Prefecture. Fishery statistics for E. pacifica have been available since 1953
from the Onagawa Fish Market.
The fishing season runs from January to June for sand lance, and from March to
April for E. pacifica. When fishing conditions for sand lance were poor the fishermen changed the net and targeted E. pacifica. An increased requirement for sea
bream culture and bait for sport fishing in the late-1960s caused an expansion in the
fishery to the north as well as along the southern coasts of Miyagi Prefecture. In the
Ibaraki Prefecture, fishing for E. pacifica began in 1972. A very large catch was
made in 1974 when, for the first time in the history of the fishery in the prefecture,
the first branch of the Oyashio Current intruded as far south as Inubo-zaki. A small
amount of E. pacifica had been landed in the southern part of Ibaraki Prefecture
previously, but large amounts were not landed until 1974. Shortly after that
Fukushima Prefecture began an E. pacifica fishery, to be followed in 1975 by Iwate
Prefecture, both becoming fishing prefectures.
Thus, the E. pacifica fishery which began in the Miyagi Prefecture developed into
an important fishery in the Sanriku (Aomori, Iwate and Miyagi Prefectures from
north to south) and the Joban (Fukushima and Ibaraki Prefectures) coastal waters
(see Fig. 3.1.4).
A small-scale fishery for E. pacifica had also been conducted along the coast of the
Japan Sea over the last 20 years in waters off the Akita, Yamagata, Shimane and
Yamaguchi Prefectures and off the eastern coast of Tsushima Islands in the Nagasaki
Prefecture (Kuroda, 1994). Catches from all of these areas have been small.
286
11.2
Krill: Biology, Ecology and Fisheries
Present status of the Euphausia pacifica fishery
The Euphausia pacifica fishery is categorised as a licensed fishery, licenses being
issued by the prefectural governor. Small boats of less than 20 t are engaged in the
fishery. One- or two-boat seines are used in all the prefectures. In the Miyagi
Prefecture, the original bow-mounted trawls are still used, but since 1991 both oneboat seine and bow-mounted trawls have been used (Kuroda & Kotani, 1994).
Photographs of these fishing boats can be seen in Nicol & Endo (1997).
A bow-mounted trawl can only catch surface swarms within 8 m of the surface. On
the other hand, one- and two-boat seines catch subsurface swarms, detected by
echo-sounder, as deep as 150 m. Over the period 1991 to 1997 in the Miyagi Prefecture, boat seines had a higher CPUE (4.7±7.8 t/boat/day) than the CPUE
achieved using bow-mounted trawls (0.3±4.0 t/boat/day). These differences being
1.5 to 15.7 times over the same fishing periods.
The fishing season lasts generally from February to July, but varies from area to
area and from year to year (Fig. 11.2). The main fishing season in Sanriku waters is
from March to April. That of Joban waters was similar until 1986, when it shifted to
later in the year, from May to June (Odate, 1991; Kuroda & Kotani, 1994), a change
which seems to reflect the warmer oceanographic condition prevailing since 1986
(Fig. 11.3). In warmer years, annual temperature minimum tends to occur later in
the year. The fishing grounds lie over the continental shelf (< 200 m) within 10±20
nautical miles of the shore. The fishing depth is usually less than 50 m in Sanriku
waters, but deeper (0±150 m) in the Joban coastal area (Kuroda & Kotani, 1994).
Fig. 11.2 Fishing period of Euphausia pacifica in Sanriku (solid line) and Joban (broken line)
waters from 1969 to 1998 (redrawn from Odate (1991) and Tohoku Natl. Fish. Res. Inst. (1998)).
Krill Fisheries in Japanese Waters
287
Fig. 11.3 Sea surface temperature (8C) in Sanriku waters (Honda, 1998). (a): interannual variation from 1981 to 1997. 1: raw data; 2: mean values in the last 17 years; 3: 12-month running mean.
(b) close-up of 1996 and 1997 with mean values in the last 17 years.
288
Krill: Biology, Ecology and Fisheries
The total annual catch of E. pacifica has increased steadily over the last 20 years
(Fig. 11.4) exceeding 40 000 t in 1978, 80 000 t in 1989 and 100 000 t in 1992. This
increase was aided in about 1975 by the introduction of plastic containers to reduce
the rate of deterioration; each of these containers holds about 30 kg of E. pacifica.
Previously, krill were stored in the fish hold in bulk. A further innovation was the
introduction of fish pumps in the 1980s (Kuroda & Kotani, 1994). In 1993, the total
catch decreased to 60 881 t, following the introduction of catch regulations which
were imposed in Miyagi and Iwate Prefectures in order to control the market and
obviate a decline in value of the catch (Kuroda & Kotani, 1994).
120 000
1.00
Joban %
100 000
Sanriku %
0.80
0.60
60 000
0.40
Percentage
Catch (ton)
80 000
40 000
0.20
20 000
0
1953 1957 1961 1965 1969 1973 1977 1981 1985 1989 1993 1997
0.00
Fig. 11.4 Annual variations in the catch of Euphausia pacifica (solid line) and the percentage
composition (bars) from Sanriku and Joban waters (redrawn from Tohoku Natl. Fish. Res. Inst.
(1998)).
11.3
Variation of annual catch of Euphausia pacifica and its
causative factors
Most of the catch in the 1970s was landed in Miyagi Prefecture. In the early to mid1980s, following the commencement of the fishery in Ibaraki Prefecture in 1972 and
in Iwate Prefecture in 1975, Ibaraki Prefecture expanded its catch taking more than
60% of the total catch in 1986. After 1988, Iwate and Miyagi Prefectures have been
the main fishing prefectures with similar catch levels (Odate, 1991; Kuroda &
Kotani, 1994); this again reflects a warmer oceanographic condition after 1988.
The fishing grounds are close to the front between the first branch of the Oyashio
Current and the coastal waters with optimal surface water temperatures of 7±98C
(Odate, 1991). Conditions suitable for fishing are therefore highly dependent on the
Krill Fisheries in Japanese Waters
289
strength of the Oyashio Current (for general current system in Japanese waters see
Endo, Chapter 3). Odate (1991) found an inverse relationship between the
southernmost latitude of the 58C sea-surface temperature isotherm and the annual
E. pacifica catch. As a more reliable indicator of the fishing condition, she calculated the area of the water mass with a temperature < 58C at 100 m depth between
358 and 428N and west of 1458E over the period 1970±1978. A very high correlation
coefficient, 0.93, was obtained between area and annual catch.
Odate (1991) classified oceanographic patterns into three types based on the E.
pacifica fishing conditions (Fig. 11.5).
O-type
C-type
D-type
42
N
Qy
Qy
Qy
40
5
5
5
15
38
15
15
36
142
15
Ku
Ku
Ku
144
142
144
142
144
Fig. 11.5 Three oceanographic conditions (O, C and D types) in Sanriku and Joban coastal waters
which relate to the formation of fishing ground for Euphausia pacifica. Figures indicate the surface
water temperature (8C) (redrawn from Odate (1991)).
(1)
O-type (good catch years, 1977, 1978, 1981, 1983±1987)
The first branch of the Oyashio Current intrudes along the Sanriku and Joban
coastal areas as far south as Kashima-nada (about 36830'N). This forms a
marked front against the warm water associated with the Kuroshio Current.
However, in extremely cold years, a water mass much colder than 58C prevails
in Sanriku coastal area, and a larger catch is landed in the Joban area compared to the Sanriku area.
290
(2)
(3)
Krill: Biology, Ecology and Fisheries
C-type (intermediate catch years, 1980, 1982, 1988, 1989)
The Oyashio Current prevails in Sanriku waters, and warm water related to the
Kuroshio Current is present in the Joban area, causing a relatively narrow
transition zone between the two currents. In this case, good catches are
restricted to Sanriku waters.
D-type (poor catch years, 1976, 1979)
The Oyashio Current is weak and only present in the northern part of Sanriku
waters. Coincidentally the Kuroshio Current is also weak leaving a very wide
transition zone between these two currents.
Kodama & Izumi (1994) undertook a similar analysis with a classification based
largely on the strength of the Oyashio Current. Their A-type is similar to the D-type
of Odate (1991). They subdivided Odate's O-type into two categories (B and C)
designated by their distance from the shore to the first branch of the Oyashio
Current. When the first Oyashio branch is strong but farther offshore, classified as
B-type, suitable water temperatures > 58C for fishing ground formation still exist in
the southern Sanriku and Joban coast, arising from which good catches are expected
in the coastal areas of Miyagi, Fukushima and Ibaraki Prefectures. When the
Oyashio first branch is very strong and comes close to the shore, classified as C-type,
suitable water temperatures > 58C are restricted to the Joban±Kashima-nada areas,
and good catches are expected only in those areas.
11.4
Prediction of fishing conditions of Euphausia pacifica
Of the various factors related to the E. pacifica fishery, the most important for the
fishermen to be able to predict are the onset and length of the fishing season, and the
area of occurrence of the fishery. At present the prediction is not carried out by
public organisations, but information on the oceanographic and fishing conditions
are provided by prefectural fisheries experimental stations in Miyagi, Fukushima
and Ibaraki (Kuroda, pers. comm.).
Kotani (1992) undertook a correlation analysis between krill catches in various
sea areas of the Pacific coast and the southernmost position of the first branch of the
Oyashio Current. A significant correlation ( p < 0.01) was found between the annual
catch in the whole area, Ibaraki and Iwate coasts, and the southernmost extent of
the first branch of the Oyashio Current. In other words, the further south the
Oyashio Current intrudes, the more catch is taken from these areas. Kotani also
found a good correlation ( p < 0.01) between the southernmost position of the first
Oyashio branch in January and the southernmost position in that year. This indicates that the krill catch can be predicted by the southernmost latitude of the first
branch of the Oyashio Current in January, about two months prior to the beginning
of the fishing season.
Kotani (1992) assumed the optimum temperature for E. pacifica swarming to be
Krill Fisheries in Japanese Waters
291
5±108C and tried to predict the first fishing day of the year by correlation analyses.
He measured the distance from the fishing grounds of Onagawa and Otsu to the
water mass with a temperature of 108C at 100 m depth for each 10-day period from
December to January and examined the relationship between this distance and the
days from the mid-day of each period to the first fishing day of the year. He found a
significant correlation ( p < 0.05) between the distance from the Onagawa fishing
grounds to the water mass and the days from mid-December and January to the first
fishing day within the Miyagi Prefecture in the year. A highly significant correlation
( p < 0.01) was also found between the distance from the Otsu fishing grounds to the
water mass and the days from late January to the first fishing day of Ibaraki Prefecture in the year.
Southward shifts of the Oyashio Current seem to depend on the wind field over
the North Pacific Ocean. Sekine (1988a, b) examined the relationship between the
North Pacific winds and the extent of southward intrusion of the Oyashio Current
and concluded that the Aleutian low pressure system was more strongly developed
and shifted southward during years of abnormal southward intrusion of the Oyashio. In relation to the El NinÄo, Sekine & Suzuki (1991) pointed out that the
anomalous southward intrusion of the Oyashio has a tendency to occur in one year
after or one year before the occurrence of the El NinÄo, when the sea-surface temperature appears warm in the west to central region of the equatorial Pacific.
Tomosada (1994) examined the relationship between atmospheric circulation
indices and water temperature at several depths in the transition zone (see Section
11.3) off northeastern Japan. He found that there is a negative correlation between
the 4th-year preceding Far East polar vortex index and the water temperature at 200
m depth in the transition zone, and a positive correlation between the 4th-month
preceding Far East zonal index and water temperature at 50 m depth in the transition zone. When the Far East polar vortex index is large, cold air flows down to
high latitudes and seawater is cooled by the cold air. When Far East zonal index is
small, the westerly wind meanders and stimulates southward flow of the cooled
water, resulting in a southward shift of the Oyashio Current. He showed the possibility of predicting long-term trends in the fishing condition of E. pacifica through
latitudinal shifts of the Oyashio Current.
Warm-core rings cut off from the Kuroshio Extension sometimes greatly affect
the fishing conditions for E. pacifica by preventing the first branch of the Oyashio
Current from intruding southward (see Endo, Chapter 3). The fishing conditions in
1990 were a good example, where a warm-core ring existed off Kinkasan Island
from January to March (Fig. 11.6). The catch of E. pacifica was 31 000 t in Iwate
Prefecture but only 10 000 t in Miyagi Prefecture, the lowest reported catch in the
last 12 years. Another example was in 1996 when a long-lived warm-core ring, 93A,
existed in the northern part of Sanriku waters and blocked the southward intrusion
of the Oyashio Current (Fig. 11.6). As a result, substantial catches were restricted to
the coastal waters of Iwate and Miyagi Prefectures, namely, the northern part of
Sanriku waters. Sekine (1988a) reported that the Tsugaru Warm Current water and
292
Krill: Biology, Ecology and Fisheries
Fig. 11.6 Two examples of the oceanographic condition in which the warm-core ring prevented
the first Oyashio branch from intruding southward and the fishing grounds of Euphausia pacifica
were restricted to the northern Sanriku waters (redrawn from the map of Tohoku Natl. Fish. Res.
Inst.).
the Kuroshio warm-core rings may block the southward intrusion of the Oyashio
Current, as apparently occurred in the years when the Aleutian low pressure system
was strongest. So not only the first branch of the Oyashio Current, but also the
presence or absence of warm-core ring(s) and the strength of the Tsugaru Warm
Current affect the fishing conditions for E. pacifica. These oceanographic conditions
must therefore be taken into account in order to predict more effectively the fishing
conditions for E. pacifica.
11.5
Management
The Euphausia pacifica fishery is designated as a licensed fishery with licenses
issued by a prefectural governor. Small boats less than 20 t are engaged in the fishery
(Table 11.1). The composition of the fleets has already been described in Section
No. boat seine
No. bow mounted trawl
Fishing period
Miyagi
No. boat seine
Fishing period
Operation time
Fishing ground
Boat size
Ibaraki
498
11 Feb±31 Jul
dawn±dusk
sea area off Ibaraki Prefecture
< 5 tonnes
283
1 Feb±31 Jul
dawn±dusk
sea area off Fukushima Prefecture
< 7 tonnes
145
208
15 Feb±31 May for boat seine*
1 Jan±31 Jul for bow mounted trawl**
until 15:00
sea area off Miyagi Prefecture
< 20 tonnes
298
all year round (until catch limit is attained)
until 15:30
sea area off Iwate Prefecture
< 20 tonnes
Contents of Licence
* The period shortened to 75 days from the first fishing day by self-regulation
** The period shortened to 1 Feb±30 Jun by self-regulation
No. boat seine
Fishing period
Operation time
Fishing ground
Boat size
Fukushima
Landing time
Fishing ground
Boat size
No. boat seine
Fishing period
Landing time
Fishing ground
Boat size
Iwate
Prefecture
Total catch 10 500 tonnes
Max. catch 6 tonnes per boat per day
Total catch 10 500 tonnes
Max. catch 6 tonnes per boat per day
Total catch 32 500 tonnes
28 500 tonnes for boat seine
4000 tonnes for bow mounted trawl
Max. number of containers per day
350 for boat > 10 tonnes
330 for boat < 10 tonnes
Total catch 28 500 tonnes
Max. number of containers per day
400 for boat > 9 tonnes
300 for boat < 9 tonnes
Regulation
Table 11.1 Contents of licence and regulation on Euphausia pacifica fishery in 1997 (Anon., 1998). A full container weighs 30 kg in Iwate and Miyagi
Prefectures and 28 kg in Fukushima and Ibaraki Prefectures.
Krill Fisheries in Japanese Waters
293
294
Krill: Biology, Ecology and Fisheries
11.2. Several hundred fishing boats are permitted to catch E. pacifica from each
prefecture with the overall number of krill fishing vessels totalling around 1500 in
the years 1994±1997. Over this period the total number of landings by all boats in
Miyagi Prefecture was 4500±6000 each season. In 1996 and 1997, there were less
than 10 landings by vessels using bow-mounted trawls, a reflection of the fact that
surface swarms did not appear in those years. In 1996, the surface-water temperature in Sanriku waters was 0.6 to 2.08C higher than the 17-year average from late
January to late March and in 1997, it was 1.5 to 4.68C higher from early January to
mid February (Honda, 1998). This suggests that the higher surface-water temperature may have prevented surface aggregations from forming.
Fishery regulations are set separately for each prefecture. The licence of the
prefectural governor defines the fishing period, the time allowed at sea before the
vessel must return to port (also referred to as the operational time), fishing area,
boat size and other factors (Table 11.1). Other regulations include total catch limit
for the season and a maximum number of plastic containers permitted for each boat
per day. Such regulations are not based on scientific rationale. Feasibility studies
have been initiated to estimate the standing stock of E. pacifica employing two
different methods. These are direct assessment using quantitative acoustics and
indirectly from egg counts in plankton surveys (see Endo, Chapter 3). There is a
clear need to develop a fisheries management regime based on a scientific assessment of the krill populations.
11.6
Economics
The price of E. pacifica on landing in 1989±1993 averaged 2.7 billion yen for the
entire area. The mean unit price per kilogram ranged from 29.6 yen in 1993 to 79.8
yen in 1996 in Miyagi Prefecture. The average unit price seems to depend on the
cumulative catch at the end of March in any year and the annual catch of the previous year.
Yoshida (1995) showed that the average unit price of E. pacifica (Y in yen per kg)
can be predicted by the cumulative catch at the end of March (X1 in 1000 t) and the
annual catch of the previous year (X2 in 1000 t) as follows:
Y= 71.1 5X1 7 0.35X2 + 99.41
with a multiple correlation coefficient of 0.89. This means that the increase in X1 by
10 000 t reduces the price by 11.5 yen, and the same increase in X2 reduces price by
3.5 yen. There is a reasonably good agreement between the real average price and
the calculated value in 1987±1994.
Availability of Antarctic krill does not seem to affect the price of E. pacifica, as
long as the harvested level of the former species does not change drastically
(Yoshida, 1995).
Krill Fisheries in Japanese Waters
11.7
295
Usage of landed Euphausia pacifica
The main commercial uses of E. pacifica are for sport fishing, feed for fish culture,
and human consumption, with the last item being by far the smallest (Kuroda &
Kotani, 1994). Kuroda and Kotani suggest that there is little competition between
Antarctic krill, E. superba and E. pacifica used for sport fishing. Whereas E. pacifica, being smaller, is used as chum bait and accounts for about 50% of the total
catch, E. superba being somewhat larger are used as a bait. The remaining 50% of E.
pacifica catch is used as an ingredient in aquaculture feeds to add reddish colour to
the skin and meat of fish such as red sea bream (Pagrus major), coho salmon
(Oncorhynchus kisutch), rainbow trout (Salmo gairdnerii), yellowtail (Seriola
quinqueradiata) and others (Kuroda, 1994), since E. pacifica contains a high level of
carotenoid pigments (5.4 mg/100 g, Sakurai & Omori 1985). As Odate (1991) has
pointed out, Japanese people favour red as a colour since it is thought to bring good
luck, consequently red sea bream and lobsters are used as offerings in celebrations.
E. pacifica is rich in essential amino acids (Kayama et al., 1976) and in n-3 series
unsaturated fatty acids such as EPA (eicosa pentaenoic acid) and DHA (docosahexaenoic acid) (Yamada, 1964), fatty acids which are known to be effective against
circulatory diseases such as arterial sclerosis (Saito, 1994). There are higher concentrations of EPA and DHA than are found in E. superba and since they constitute
more than 30% of the total fatty acids, this is likely to ensure that E. pacifica
continues to be an important food source for humans.
11.8
Other Euphausiid species commercially exploited in the
Japanese waters
Thysanoessa inermis has been commercially exploited since the early 1970s in the
inshore waters of Shakotan Peninsula and Yagishiri Island, off the western coast of
Hokkaido (Fig. 11.7, Hanamura et al., 1989; Kotori, 1994). Surface aggregations of
this species are fished in the daytime usually from early March to early April. A
spoon net, with a diameter of 2 m and a 3±4 m handle, is used to target the
aggregations. The yearly catch varies from very little up to 200 t. The price has
varied from 75 to more than 3000 yen per kg and was reasonably stable at around
100 yen per kg, in 1994 and 1995. There was no catch in 1996 (Kotori, pers. comm.).
The T. inermis fishery is categorised as a recognised fishery by the Fisheries
Regulation Committee of Ishikari-Shiribeshi waters. There are no regulations
affecting the catch, although the fishing period is limited to February±June. Biological characteristics of the aggregating individuals have been investigated by
Hanamura et al. (1989) and the aggregations have been shown to comprise fully
mature individuals with males possessing spermatophores and females having
spermatophores attached to the thelycum. Hanamura et al. concluded that T.
inermis were engaged in reproductive activities during the swarming season.
296
Krill: Biology, Ecology and Fisheries
140 E
142
144
Rishiri Is.
SEA OF
OKHOTSK
45 N
SEA OF
JAPAN
Yagishiri Is.
44
HOKKAIDO
43
Shakotan
Pen.
SEA OF
JAPAN
40 N
42
PACIFIC
OCEAN
130 E
30
140
Fig. 11.7 Location of fishing grounds (hatched area) for Thysanoessa inermis off the western coast
of Hokkaido (Hanamura et al., 1989).
`Ami-ebi' or E. nana, an allied species of E. pacifica, has been commercially
exploited for 26 years in Uwajima Bay, Ehime Prefecture, Shikoku (Fig. 11.8,
Hirota & Kohno, 1992). Fishery statistics have been available since 1976. Over the
period 1980 to 1991, the yearly catch varied from 2000 to 5000 t. Since this period,
there have been no substantial catches reported (Hirota, pers. comm.). Two fishing
methods are employed to catch E. nana aggregations. One is purse seine operated at
night during the period March to July by a group of vessels comprising a netting
boat, a transport-boat, and three or less light-boats using fish gathering lamps. These
seines usually target sardines and mackerel, but catch E. nana when the conditions
for catching these species are poor. The other fishing method uses a boat seine of
Krill Fisheries in Japanese Waters
297
33 20'N
UWAJIMA
CITY
UWAJIMA
BAY
34 N
33 10'N
32 N
132 20'
E
132 30'
E
130
E
135
E
Fig. 11.8 Map of the Uwajima Bay area, Ehime Prefecture where Euphausia nana is commercially exploited (adapted from Hirota & Kohno (1992)).
drawers type operated in the daytime by a troop which consists of two netting boats,
a boat with fish sounder and a transport-boat, from spring to early summer.
The catch of E. nana is used mainly as a feed in the aquaculture industry for the
production of red sea bream. The price is about 50 yen per kg. When supply exceeds
demand the price falls, but, to protect the fishing industry the catch is then controlled taking into account the E. pacifica catch in Sanriku and Joban waters.
The distribution of E. nana is neritic with a range extending from the southern
Japanese coasts to the East China Sea and Taiwanese waters (Hirota & Kohno,
1992). The species does not seem to be endemic to Uwajima Bay because the fishing
season is restricted and adults are reported to perform diel vertical migration with
daytime depth of 300±400 m in Sagami Bay (Hirota et al., 1983).
References
Anon. (1998) Fishing condition of Euphausia pacifica in Sanriku and Joban coastal waters in
1997. Rep. Res. Meeting on Krill Resources and Oceanogr. Conditions in the Northwest.
Pacific 57±62 (in Japanese).
Hanamura, Y., Kotori, M. & Hamaoka, S. (1989) Daytime surface swarms of the euphausiid
Thysanoessa inermis off the west coast of Hokkaido, northern Japan. Mar. Biol. 102, 369±
76.
Hirota, Y. & Kohno, Y. (1992) A euphausiid fishery in Uwajima Bay, Ehime Prefecture.
298
Krill: Biology, Ecology and Fisheries
Fisheries Biology and Oceanography in the South-western Waters of Japan 8, 89±95 (in
Japanese).
Hirota, Y., Nemoto, T. & Marumo, R. (1983) Vertical distribution of euphausiids in Sagami
Bay, central Japan. La mer 21, 151±63.
Honda, O. (1998) Characteristic features in the surface and mid water temperatures in
Tohoku waters. Rep. Res. Meeting on Krill Resources and Oceanogr. Conditions in the
Northwest Pacific 32±41(in Japanese).
Iwate Prefecture (1999) Fisheries of Iwate Prefecture. 31 pp. (in Japanese).
Kayama, M., Ikeda, Y. & Komaki, Y. (1976) Studies on the lipids of marine zooplankton, with
special reference to wax ester distribution in crustaceans and its in vivo formation.
Yukagaku (Lipid Chemistry) 25, 329±34 (in Japanese with English abstract).
Kodama, J. & Izumi, Y. (1994) Factors relevant to the fishing ground formation of Euphausia
pacifica and the relation to the demersal fish resources. Gekkan Kaiyo (Kaiyo Monthly) 26,
228±35 (in Japanese).
Kotani, Y. (1992) What aspects and how to predict fishing conditions of Euphausia pacifica?
Rep. Res. Meeting on North Pacific Krill Resources No. 1, 108±113 (in Japanese).
Kotori, M. (1994) A euphausiid fishery in the west coast of Hokkaido. Gekkan Kaiyo (Kaiyo
Monthly) 26, 248±50 (in Japanese).
Kuroda, K. (1994) Euphausiid fishery in the Japanese waters and related research activities.
Gekkan Kaiyo (Kaiyo Monthly) 26, 203±209 (in Japanese).
Kuroda, K. & Kotani, Y. (1994) Euphausiid fishery and interannual variation of the fishing
conditions in the Sanriku-Joban coastal areas. Gekkan Kaiyo (Kaiyo Monthly) 26, 210±217
(in Japanese).
Miyagi Prefecture (1998) White book on fisheries in Miyagi Prefecure. 74 pp. (in Japanese).
Nicol, S. & Endo, Y. (1997) Krill fisheries of the world. FAO Fish. Tech. Paper 367, 1±100.
Odate, K. (1991) Fishery biology of the krill, Euphausia pacifica, in the Northeastern Coasts
of Japan. Suisan Kenkyu Sosho (Library of Fisheries Study) 40, 1±100 (in Japanese with
English abstract).
Saito, H. (1994) Element composition of euphausiids, with special reference to lipids. Gekkan
Kaiyo (Kaiyo Monthly) 26, 224±28 (in Japanese).
Sakurai, S. & Omori, A. (1985) On the body color of Euphausia pacifica. Business Rep. of the
1985 fiscal year, Ibaraki Pref. Fish. Exp. St., 169 (in Japanese).
Sekine, Y. (1988a) Anomalous southward intrusion of the Oyashio east of Japan, 1. Influence
of the seasonal and interannual variation in the wind stress over the North Pacific. J.
Geophys. Res. 93, 2247±55.
Sekine, Y. (1988b) A numerical experiment on the anomalous southward intrusion of the
Oyashio east of Japan, Part I. Barotropic model. J. Oceanogr. Soc. Japan 44, 60±67.
Sekine, Y. & Suzuki, Y. (1991) Occurrence of anomalous southward intrusion of the Oyashio
and the variation in global atmospheric circulation. Umi to Sora (Ocean and Sky) 67, 11±23
(in Japanese with English abstract).
Tohoku National Fisheries Research Institute (1998) Interannual variation in the catch and
fishing period of Euphausia pacifica. Rep. Res. Meeting on Krill Resources and Oceanogr.
Conditions in the Northwest. Pacific 51±56 (in Japanese).
Tomosada, A. (1994) A feasibility study of predicting water temperature variation in the
Transition area off northeastern Japan. Rep. Res. Meeting on North Pacific Krill Resources
No. 3, 100±101 (in Japanese).
Krill Fisheries in Japanese Waters
299
Yamada, M. (1964) On the lipids of plankton. Bull. Jap. Soc. Scient. Fish. 30, 673±81 (in
Japanese).
Yoshida, H. (1995) A study on the price formation mechanism of Euphausia pacifica. Bull.
Jap. Soc. Fish. Oceanogr. 59, 36±38 (in Japanese).
Chapter 12
Management of Krill Fisheries in the
Southern Ocean
Denzil Miller and David Agnew
12.1
Introduction
Over the past thirty years, the growing demand for protein from the sea along with
increasingly restricted access to historical fishing grounds has resulted in the
development of many `unconventional' fisheries (Robinson, 1982; Budzinski et al.,
1985; Miller, 1990). The Antarctic krill (Euphausia superba Dana, hereinafter
referred to as `krill' in this chapter) fishery is a clear example of such development.
Exploitation of Antarctic marine living resources has been characterised by
intense and sporadic cycles (Gulland, 1977, 1983) often resulting in severe depletion
of harvested stocks (especially seals, whales and finfish) (Laws, 1977, 1985, 1989;
Kock, 1991, 1992). This has raised serious concerns about the future effective
management and sustainable utilisation of resources (Mitchell & Sandbrook, 1980)
with krill, as a key species, assuming prominence (Nicol, 1989; Miller, 1991).
Krill has long been recognised as an integral component of many Antarctic
marine food webs (e.g. Marr, 1962; Everson, 1977, 1984; Knox, 1984; Miller &
Hampton, 1989 among others). Several important attributes enhance its potential as
an exploitable resource (Everson, 1977, 1978; El-Sayed, 1988; Nicol, 1989) as well as
its `availability' to many predators. These are, inter alia, high global abundance
(Everson, 1977; Gulland, 1983), high nutritional value (Grantham, 1977; Budzinski
et al., 1985), and a tendency to aggregate which increases the ease of location and
capture (Eddie, 1977; El-Sayed & McWhinnie, 1979).
Exploratory fishing for krill commenced in the early 1960s with catch levels being
initially low (Table 12.1). The build-up of catches was slow and it was not until the
1973±1974 season that the fishery assumed commercial significance. In this review,
catch statistics are used to indicate areas where the fishery has been most active and
from where the largest catches have been taken. Trends in catches are analysed to
provide insights into possible relationships between the fishery, its operational
constraints and krill distribution/abundance (i.e. availability). Earlier reviews by
Miller (1990) and Agnew & Nicol (1997) are updated to include recent efforts by the
23-nation Commission for the Conservation of Antarctic Marine Living Resources
(CCAMLR) and its Scientific Committee (SC-CAMLR) to develop a scientifically
300
Krill Fisheries in the Southern Ocean
Table 12.1
1990).
301
Annual krill catches prior to 1973/74 (from Everson, 1978; Bengtson, 1984; Miller,
Season
Catch (t)
1961/62
1963/64
1964/65
1966/67
1968/69
1969/70
1970/71
1971/72
1972/73
4
70
306
?
> 140
100
1300
2100
7459
Comments
Reference
Krill
Krill
Krill
Krill
Krill
UMC*
UMC*
UMC*
Krill
Burukovskiy & Yaragov (1967)
Stasenko (1967)
Nemoto & Nasu (1975)
Nemoto & Nasu (1975)
Ivanov (1970)
FAO (1976)
FAO (1976)
FAO (1976)
FAO (1976)
Nemoto & Nasu (1975)
* Catches of `Unspecified Marine Crustacea' for areas closely adjacent to the Antarctic area assumed by
Everson (1978) to be krill.
based management regime for the krill fishery. Advice on the management of
Antarctic marine living resource exploitation has been developed by specialist
working groups of SC-CAMLR. For krill these comprised the Working Group on
Krill (WG-Krill) (1987±1994) and the Working Group for the CCAMLR Ecosystem
Monitoring Programme (WG-CEMP) (1986±1994). In 1994 the two Groups were
combined into the Working Group for Ecosystem Monitoring and Management
(WG-EMM).
12.2
Trends in the fishery
Available data
Krill catches are reported from three major statistical areas in the Antarctic, which
are each divided into a number of subareas or divisions (Fig. 12.1). Summary catch
and effort data are submitted annually to CCAMLR, with catches being attributed
to split-years ± a split-year being the twelve-month period from 1 July in one year to
30 June in the next. Since 1990, CCAMLR fisheries statistics have been published
annually in a Statistical Bulletin (Statistical Bulletins 1990±1998). The former Soviet
Union was responsible for the bulk of catches taken during the period under review.
Soviet catch data in all areas were only separated by month from 1983 and effort
data (e.g. hours fished) are only available from that year. From 1991±1992, krill
catches and the reporting thereof by the former Soviet Union were split between the
Russian Federation (a CCAMLR Member) and Ukraine (non-CCAMLR Member
prior to 1995). For the bulk of the analyses post-1992, we have combined the catches
of these two nations, considering them synonymous with those of the Soviet Union.
The data used to investigate relationships between sea-ice and the krill fishery
were derived from ice concentration data held in the database of the National Snow
302
Krill: Biology, Ecology and Fisheries
Fig. 12.1 Statistical Areas, Subareas and Divisions of the CAMLR Convention Area (from
Anon., 1995). With permission from the CCAMLR.
and Ice Centre, Colorado, USA and from the CCAMLR Statistical Bulletin
(Statistical Bulletins 1990±1998). The possible association between sea-ice and krill
catch-per-unit effort (CPUE) was investigated by comparing ice cover in October of
one split-year with CPUE in that split-year. For example, ice cover in October 1988
(i.e. from the 1988±1989 split-year) was compared with krill CPUE in 1988±1989. In
Subareas 48.1 and 48.2, most fishing takes place in summer and autumn (see Section
on Seasonal distribution (i.e. 3±6 months after the October calculation date)). In
Subarea 48.3, fishing takes place in the winter (June±September) and so the correlation was performed between ice in October and krill CPUE of the following
split-year (e.g. ice in October 1988 from the 1988±1989 split-year was compared with
CPUE from the 1989±1990 split-year).
Catch trends
Annual catches
From the first commercial takes in the early 1970s, krill catches rose steadily from 19
785 t in 1973±1974 to a peak of 528 201 t in 1981±1982 (Fig. 12.2). Catches then
declined sharply until 1983±1984 due to marketing and processing problems possibly
associated with the discovery of high levels of fluoride in the exoskeleton of krill
0
100000
200000
300000
400000
1986/87
1985/86
1984/85
Split year
Others
Jpn
Pol
1997/98
1996/97
1993/94
1992/93
1991/92
1990/91
1989/90
1988/89
1987/88
1983/84
1982/83
1981/82
1980/81
1979/80
1978/79
1977/78
1976/77
1975/76
1974/75
1973/74
Fig. 12.2 Annual krill catches in tonnes from 1973/74 to 1997/98 by nation. Years are split years (see text for explanation), JPN =
Japan, POL = Poland, SUN = Soviet Union (including Russia and Ukraine post-1990/91) and OTHERS = Other krill fishing
nations (Bulgaria, Chile, Korea and United Kingdom).
Tonnes
Sun
1994/95
500000
1995/96
600000
Krill Fisheries in the Southern Ocean
303
304
Krill: Biology, Ecology and Fisheries
(Nicol & Endo, 1997; Ichii, Chapter 9). Following the solution of these technical
problems by rapid processing and the development of improved peeling machines,
catches increased again, levelling off at around 300 to 400 000 t. The next major
impact on the krill fishery was triggered by the breakup of the Soviet Union in 1991
with the subsequent removal of fuel subsidies for Russian and Ukrainian vessels
(Nicol & Endo, 1997). The subsequent drop in catches is graphically illustrated in
Fig. 12.2.
Other key nations (mainly Poland, Chile and Korea) have accounted for about
4% of the total catch. Korea entered the fishery in 1978±1979, fishing sporadically in
Subarea 58.4 from 1978±1979 to 1983±1984 and took about 9000 t during that time.
It re-entered the fishery in 1986±1987, moving its operations to Subareas 48.1 and
48.2, and taking about 1500 t annually until 1991±1992. Both Chile and Poland
undertook exploratory fishing for krill between 1976 and 1977, but did not enter the
fishery in earnest until 1982±1983 and 1985±1986 respectively. Chile built up to
annual catches of about 5000 t throughout the 1980s, leaving the fishery after the
1993±1994 season. Poland has taken variable catches between 2000 and 20 000 t to
date. The United Kingdom (UK) has been a recent entry into the fishery taking 308
and 634 t in 1996±1997 and 1997±1998 respectively. Very recently there has been a
resurgence of interest in fishing for krill, with the entry of some new players such as
the UK, Argentina and USA (CCAMLR, 1998; SC-CAMLR, 1998, 1999), with
some interest also being shown by Canada, a Convention Acceding State but not a
Member of CCAMLR.
Initially, the krill fishery was distributed in all three CCAMLR Statistical Areas
(48, 58 and 88; Fig. 12.1). Japan started fishing in Division 58.4.2 which was logistically convenient. However, following comprehensive exploration of most areas
around the Antarctic continent, the fishery became concentrated in the Atlantic
sector (Area 48) where the most predictable concentrations of krill are to be found
(Fig. 12.3). Catches in Areas 58 and 88 were located along the continental shelf
break, relatively close to the Continent (Ichii, 1988, 1990). Catches in Area 48 have
also been concentrated on shelf breaks, to the north and west of the South Shetland
Islands, in a rather more diffuse area north of the South Orkney Islands and to the
north of South Georgia (Agnew 1993; Ichii, Chapter 9).
There has been one reported catch of krill outside the CCAMLR Convention
Area in 1992±1993 by Poland immediately north of Subarea 48.1 (FAO Statistical
Division 41.3.2) just to the south of the Antarctic Polar Front (Fig. 12.1). Occasionally patches of krill have been reported to drift in a north-easterly direction
from Elephant Island to north of 608S (i.e. out of the CCAMLR Area) with catches
ultimately being made in the vicinity of 57±608S, 50±608W (Sushin & Myskov, 1993).
Drifter buoys released close to the South Shetland Islands have also been observed
to transit Division 41.3.2 en route to South Georgia (Ichii et al., 1993), while the First
International BIOMASS (Biological Investigations of Marine Antarctic Systems
and Stocks) Experiment (FIBEX)(Anon., 1986) found significant quantities of krill
in the same region (Trathan et al., 1998). Division 41.3.2 thus appears to be bio-
0
100 000
200 000
300 000
400 000
1986/87
1985/86
1984/85
Split year
Area 58 + 88
Unspec Area 48
Subarea 48.3
1994/95
1993/94
1992/93
500 000
Subarea 48.2
1997/98
1996/97
1991/92
1990/91
1989/90
1988/89
1987/88
1983/84
1982/83
1981/82
1980/81
1979/80
1978/79
1977/78
1976/77
1975/76
1974/75
1973/74
Fig. 12.3 Annual krill catches in tonnes by CCAMLR Statistical Area or Subarea. Years are split years (see text for explanation), A58&88 = Areas 58 and 88 combined, Unspec area 48 = Unspecified locations within Area 48, S48.1 = Subarea 48.1, S48.2 =
Subarea 48.2 and S48.3 = Subarea 48.3.
Tonnes
Subarea 48.1
1995/96
600 000
Krill Fisheries in the Southern Ocean
305
306
Krill: Biology, Ecology and Fisheries
geographically contiguous with Subarea 48.1 and could be considered as part of the
Scotia Sea system, along with Subareas 48.1, 48.2 and 48.3.
Seasonal distribution
The seasonal pattern of krill catches around the Antarctic is, as one would expect,
critically linked to seasonal sea-ice distributions. In areas adjacent to the Antarctic
Continent fishing generally commences in the early austral spring as the ice edge
moves southwards exposing the krill-rich waters of the shelf break, and ceases in
about May with krill appearing to remain over the shelf break under the ice during
the winter (Ichii, 1990). In Subarea 48.2, which is slightly more oceanic than other
close-continent areas, fishing commences at a similar time but catches are taken
over a longer period (January to May) and usually peak slightly later than in Subarea 48.1 (March±April). By contrast, the waters around South Georgia (Subarea
48.3) remain ice-free in winter, and so constitute the only waters Antarctic-wide
which contain fishable krill concentrations at this time of year. The positions of krill
catches in the Atlantic sector have therefore been clearly linked to the progressive
encroachment of ice northwards during the autumn and winter so forcing the fishing
fleet also to move north (Dolzhenkov et al., 1989; Miller, 1990; Everson & Goss,
1991; Agnew, 1993).
There has been a general trend over the last decade for fishing in Subareas 48.1
and 48.2 to show a shift in median catch date to later in the season (Fig. 12.4). In
48.1, the fishing period extended from about 4 months (December±March) in the
early 1980s to 8 months (November±June) now. However, in 48.2, although it
expanded from 6 months (December to May) in the early 1980s to 9 months in the
early 1990s, the fishing season has since contracted to about 4 months (March±
June). In 48.3, by contrast, the median catch date (of the winter fishery) has
remained relatively unchanged, and as in Subarea 48.2 the duration of the fishery
has contracted from 10 (January±October) to 5 months (May±September).
There are a number of explanations for these observed changes in fishing pattern,
both environmental and operational. In Subarea 48.2, the trend in median catch
date may have been a response to increasing availability of ice-free areas late in the
season, allowing fishing to continue to June in many years. The number of ice-free
days around Signy Island (Subarea 48.2), calculated by SC-CAMLR (1996) as the
duration from ice retreating in the spring to ice returning in the autumn, displays a
well-documented cyclic pattern (Murphy et al., 1995), but also shows a general trend
of increasing duration in the ice-free period over the last decade (Fig. 12.5). In
contrast, oceanographic conditions around the Antarctic Peninsula mean that
although the cyclic phenomenon is still apparent in Subarea 48.1 there is little longterm trend. With the exception of the period 1981±1982 to 1984±1985, no clear
linking of the later median catch date to the availability of ice-free areas is possible
(Figs 12.4 and 12.5). Furthermore, Agnew (1993) has noted the close dependence of
the distribution of catches in Subarea 48.2 on variable sea-ice conditions, in contrast
Krill Fisheries in the Southern Ocean
307
Fig. 12.4 Median catch date by month. For krill fisheries in CCAMLR Subareas 48.1 and 48.2 the
median date refers to the summer fishery of the split year. For Subarea 48.3, the median date refers
to the winter fishery which commenced on 1 January of a split year. For instance, for 1986±87 this is
the winter fishery of the year 1987, taking place between 1 January 1987 and 31 December 1987.
48.1 = Subarea 48.1, 48.2 = Subarea 48.2 and 48.3 = Subarea 48.3.
to Subarea 48.1 where a remarkably consistent pattern of catches from year to year
was found. The trend to later autumn fishing in Subareas 48.1 and 48.2 may
therefore only partially be driven by decreasing ice coverage (see Section 12.5).
12.3
Developing a management regime
Background
The CAMLR Convention was negotiated between 1978 and 1980, entering into
force in 1982. The substantial krill fishery throughout this time, coupled with
growing interest from countries such as Poland, Korea and Chile, imparted an
urgency to CCAMLR's establishment. As already indicated, this was particularly
attributable to krill's perceived pivotal role as a key ecosystem component and
concern that it not be over-exploited (Edwards & Heap, 1981). Therefore, on its
entry into force CCAMLR was faced with the daunting task of developing an
effective strategy to meet the Convention's primary objectives and to manage an
anticipated expansion in krill exploitation levels (Everson, 1988a; Miller, 1991).
The fundamental management objectives of CCAMLR are set out in Article II of
the Convention (Table 12.2). From the outset the ecological importance of krill as a
key species in the Antarctic marine ecosystem, and the consequent need to manage
Fig. 12.5 Number of ice-free days (duration of the period between ice retreat and ice return) calculated by SC-CAMLR (1997)
for sites in Subareas 48.1 and 48.2.
308
Krill: Biology, Ecology and Fisheries
Krill Fisheries in the Southern Ocean
309
Table 12.2 Article II of the CAMLR Convention (from Anon., 1995).
(1)
The objective of this Convention is the conservation of Antarctic marine living
resources
(2)
For the purposes of this Convention, the term `conservation' includes rational use
(3)
Any harvesting and associated activities in the area to which this convention applies shall
be conducted in accordance with the provisions of this Convention and with the following
principles of conservation:
(a)
prevention of decrease in the size of any harvested population to levels below those
which ensure its stable recruitment. For this purpose its size shall not be allowed to
fall below a level close to that which ensures the greatest net annual increment;
(b) maintenance of the ecological relationships between harvested, dependent and
related populations of Antarctic marine resources and the restoration of depleted
populations to the levels defined in sub-paragraph (a) above, and
(c) prevention of changes or minimization of the risk of change in the marine
ecosystem which are not potentially reversible over two or three decades, taking
into account the state of available knowledge of the direct and indirect impact of
harvesting, the effect of the introduction of alien species, the effects of associated
activities on the marine ecosystem and of the effects of environmental changes,
with the aim of making possible the sustained conservation of Antarctic marine
living resources.
stocks in such a way as to minimise potential ecological risks both to krill and to krill
predators, was recognised (Everson, 1977; Knox, 1984; Butterworth, 1986; Miller,
1991; Nicol & de la Mare, 1993). CCAMLR was the first international fisheries
arrangement to explicitly assume both precautionary and ecosystem approaches as
basic management principles. As a result, CCAMLR has addressed management of
the krill fishery from two perspectives: a bottom-up, `direct', approach (i.e. examining krill as exploitable populations with predictable dynamics), and a top-down,
`indirect', approach (i.e. considering the potential effects of krill exploitation on the
ecosystem as a whole).
Formulating operational management objectives
Butterworth (1986) was among the first to recognise that more traditional
approaches to fisheries management, such as F0.1 are unlikely to be applicable to
krill. This is attributable to such factors as potential density-dependent fishing
mortality arising from the species' aggregating behaviour, problems associated with
accurately ageing krill and the general inapplicability of F0.1 in multi-species
applications. F0.1 is often used as a reference point in fisheries management. If
unexploited fisheries are subject to gradual increases in fishing mortality (F) the
yield (catch) per individual recruit rises to either a maximum peak or asymptote F0.1
which is that fishing mortality at which the increase in yield-per-recruit is 10% of the
increase in yield-per-recruit at the origin. As a management reference point it is at a
lower F, which is more conservative than the often quoted Fmax. Subsequently the
310
Krill: Biology, Ecology and Fisheries
need for `operational management principles' (after Butterworth, 1990) to manage
krill stocks in specified `management' areas was recognised by CCAMLR. Four
inter-related elements were identified as important (SC-CAMLR 1990):
(1)
(2)
(3)
(4)
Requiring a basis for assessing the status of krill stocks (an estimator).
Developing an algorithm to specify appropriate regulatory procedures [i.e. a
`Catch control law' such as a Total Allowable Catch (TAC)].
Simulating and testing the performance of management procedures [i.e.
components of both (1) and (2)] and
Developing operational definitions of Article II (i.e. quantities measurable
from field observations) to provide criteria against which management performance procedures may be assessed.
The operational management principles accepted by CCAMLR in the development
of krill management goals (SC-CAMLR, 1990; CCAMLR, 1990) were:
(1)
(2)
(3)
(4)
To aim to keep krill biomass at a level higher than would be the case for singlespecies harvesting considerations.
Given that krill dynamics have a stochastic component, to focus on the lowest
biomass that might occur over a future period, rather than on the average
biomass at the end of that period, as might be the case in a single-species
context.
To ensure that any reduction of food to predators which may arise out of krill
harvesting is not such that land-breeding predators with restricted foraging
ranges are disproportionately affected compared to predators in pelagic
habitats and
To examine what levels of krill escapement are sufficient to meet the
reasonable requirements of predators.
These principles are applied so as to account for sustained consistency in catch
levels with time while taking into account uncertainties (i.e. krill demand of predators) so that, on available information, at least, the possibility of violating the
objectives of Article II is reduced (Butterworth et al., 1994).
In a practical sense, CCAMLR's first task was to identify management areas in
which to apply the above principles. However, the delineation of krill stock
boundaries has proved extremely difficult since the extent to which krill are resident
in various areas, or move between them as passive particles in the eastward flowing
Antarctic Circumpolar Current, is largely unknown and likely to be quite variable
(Miller & Hampton, 1989; see Section 12.5, Krill flux). For practical purposes,
therefore, CCAMLR has focused its efforts on areas where the fishery is primarily
located (i.e. Subareas 48.1, 48.2, 48.3, and to a lesser degree 58.4). These areas,
particularly 48.1 and 48.3, contain sites where large colonies of land-based, krill
predators breed or are located (Bengtson, 1984; Croxall et al., 1988; Croxall, 1990).
Krill Fisheries in the Southern Ocean
311
Consequently, all elements of the management paradigm (fishery, krill and landbased predators) overlap in space and/or time to varying degrees.
Second, methods for estimating appropriate levels of harvesting (i.e. yield) were
needed. A number of different approaches have been developed for this purpose ±
both direct methods of estimating yield, based on surveys of stock biomass, and
indirect methods of estimating trends in biomass. Finally, the effect of krill harvesting on dependent predators has also been incorporated at various levels, with
explicit account being taken of such effects in the yield models in association with
subsequent monitoring of potential changes likely to be associated with krill harvesting. The following sections describe these various approaches.
12.4
Estimating stock size and yield
Direct approaches
Estimation of krill yield
At the outset, CCAMLR recognised that a substantial amount of data would be
required before traditional feedback assessments might be used for krill. Given the
size of most management areas for the species and difficulties inherent to determining age (Siegel & Nicol, Chapter 5), it is doubtful whether traditional annual
assessments of krill will ever be achievable (Butterworth, 1986; Everson, Chapter 3).
CCAMLR therefore concentrated on developing calculations of long-term yield
based on the approach of Beddington & Cooke (1983), as initially modified by
Butterworth et al. (1992, 1994). In this approach, estimates of krill recruitment
variability, growth and natural mortality are used in a stochastic simulation to
determine the effects of various levels of harvesting on the population. For each
level of harvesting, the parameter g in the equation Y = gBo is calculated and the
stock is tracked over a 20-year period (where Y = yield and Bo = unexploited
biomass). Because this stochastic model is run a large number of times, the distribution of various management quantities can be determined and the probability
of various outcomes of both management and fisheries-associated actions can be
calculated. To select an appropriate level of yield, a three part decision rule,
described in Table 12.3, was developed to select an appropriate value for g.
This approach requires an estimate of pre-exploitation biomass (Bo) from
acoustic survey information. In its first application (SC-CAMLR, 1991), acoustic
estimates of krill biomass from the 1981 FIBEX survey (Anon., 1986) were used to
estimate Bo in Statistical Area 48. The FIBEX estimate was favoured since it covered Area 48 as a whole and there was little necessity for adjustment to account for
krill flux (i.e. movement). A year after initiating the above, CCAMLR adopted its
first krill Conservation Measure (CM)(CM 32/X)(Table 12.4) in 1991 using results
from the yield model (g was estimated as 0.1 and the FIBEX Bo estimate for Area 48
was 15 million t).
312
Krill: Biology, Ecology and Fisheries
Table 12.3 Three-part decision rule to select value of g (proportionality coefficient) used by
CCAMLR in setting precautionary catch limits for the Antarctic krill fishery (SC-CAMLR, 1994).
See text for full details of management procedures and their development.
(i)
g (g1) is chosen so that probability of spawning biomass dropping below 20% of preexploitation median level over a 20-year harvesting period is 10%;
(ii)
g (g2) is chosen so that median krill escapement over a 20-year period is 75%, and
(iii) The lower of g1 and g2 is selected as the level of g for the calculation of krill yield.
CM 32/X incorporated two `limitations' on the uncontrolled expansion of krill
fishing. Firstly, the precautionary krill catch limit of 1.5 million t for Area 48 was
seen as being sufficiently above prevailing historic catch levels to allow for
reasonable growth of the fishery, but low enough to minimise the possibility of a
detrimental impact on krill stocks (i.e. some allowance was made for uncertainty in
parameter estimates used in the yield model) (SC-CAMLR, 1991). Secondly, the
`catch trigger' (620 000 t) for Subareas 48.1, 48.2 and 48.3 was a little higher than the
largest annual krill catch to date. This was perceived to be the level at which rapid
expansion of catch is most likely so necessitating suitable subdivision of catch by
Subarea (i.e. paragraph 3 of CM 32/X) to avoid possible unacceptable concentration
of catch within the foraging areas of vulnerable predators (see below and Section
12.6, Overlap of krill fishery and predators)(SC-CAMLR, 1991).
Initially the krill yield model used a fixed natural mortality, M. However, there is
considerable uncertainty concerning a suitable level for this parameter, so in later
models it was ascribed to a uniform distribution between 0.6 and 1.0. Other parameters, such as variable age-at-recruitment, seasonal growth and catch history, as
well as age-at-first-capture were also added to the model. Prior distributions for a
number of parameters were refined, particularly recruitment variability and the
relationship between this and likely variability in M (de la Mare 1994a, b).
Recruitment variability is now estimated from survey data as variation in the ratio
of 1- and 2-year-old krill in the older population. There is a considerable time series
of surveys from which this parameter can be estimated, especially around the
Antarctic Peninsula region (Siegel et al., 1998).
Recently an age-based krill population model has been constructed for the
Antarctic Peninsula. This confirms that the recruitment, growth and natural mortality estimates used in the general yield model are internally consistent and are
capable of reproducing population trends similar to those observed in surveys
(Murphy et al., 1999). The effects of serial correlation in krill recruitment have also
been investigated, especially following observations of linkages between recruitment and sea-ice cover (Loeb et al., 1997), and cyclicity in sea-ice dynamics (Murphy
et al., 1995). Finally, Bo values were recalculated for various Statistical Subareas
(Trathan et al., 1993, 1995) using a revised acoustic target strength for krill (after
Everson et al., 1990; Foote et al., 1990; Greene et al., 1990). Subsequently, CCAMLR
Krill Fisheries in the Southern Ocean
313
Table 12.4 CCAMLR Conservation Measures for krill.
CONSERVATION MEASURE 32/X
The total catch of Euphausia superba in Statistical Area 48 shall be limited to 1.5 million
tonnes in any fishing season. A fishing season begins on 1 July and finishes on 30 June of the
following year.
This limit shall be kept under review by the Commission, taking into account the advice of the
Scientific Committee.
Precautionary limits to be agreed by the Commission on the basis of the advice of the Scientific
Committee shall be applied to subareas, or on such other basis as the Scientific Committee
may advise, if the total catch in Subareas 48.1, 48.2, and 48.3 in any fishing season exceeds
620 000 tonnes.
For the purposes of implementing this Conservation Measure, the catches shall be reported to
the Commission on a monthly basis.
*CONSERVATION MEASURE 45/XI
The total catch of Euphausia superba in Statistical Division 58.4.2 shall be limited to 390 000
tonnes in any fishing season. A fishing season begins on 1 July and finishes on 30 June of the
following year.
This limit shall be kept under review by the Commission, taking into account the advice of the
Scientific Committee.
For the purposes of implementing this Conservation Measure, the catches shall be reported to
the Commission on a monthly basis.
CONSERVATION MEASURE 45/XIV
The total catch of Euphausia superba in Statistical Division 58.4.2 shall be limited to 450 000
tonnes in any fishing season. A fishing season begins on 1 July and finishes on 30 June of the
following year.
This limit shall be kept under review by the Commission, taking into account the advice of the
Scientific Committee.
For the purposes of implementing this Conservation Measure, the catches shall be reported to
the Commission on a monthly basis.
**CONSERVATION MEASURE 46/XI
If the total catch of Euphausia superba in Statistical Subareas 48.1, 48.2 and 48.3 in any fishing
season exceeds 620 000 tonnes, then catches in the following Statistical Subareas shall not
exceed the precautionary catch limit prescribed below:
Antarctic Peninsula
South Orkney Islands
South Georgia
South Sandwich Islands
Weddell Sea
Bouvet Island region
Subarea
Subarea
Subarea
Subarea
Subarea
Subarea
48.1
48.2
48.3
48.4
48.5
48.6
420 000
735 000
360 000
75 000
75 000
300 000
tonnes
tonnes
tonnes
tonnes
tonnes
tonnes
Notwithstanding these subareal limits, the total sum of catches in any fishing season in all
Subareas shall not exceed the precautionary catch limit of 1.5 million tonnes for the whole of
Statistical Area 48 prescribed by Conservation Measure 32/X. A fishing season begins on 1
July and finishes on 30 June of the forthcoming year.
The above precautionary catch limits shall apply to the fishing seasons 1992/93 and 1993/94
after which time they will be reviewed by the Commission, taking into account the advice of
the Scientific Committee.
* revised in 1995
** lapsed in 1994
314
Krill: Biology, Ecology and Fisheries
has agreed to re-survey krill Bo in Area 48 during the austral summer of 1999±2000
(CCAMLR, 1998).
Alongside development of the above model, two important additional management principles were elaborated. Firstly, CCAMLR continually requested consideration of the possible allocation of the area precautionary catch limit of 1.5
million t to subareas within Area 48. A variety of methods were identified (e.g.
based on historic catch levels, predator needs and various proportionate combinations of results from the yield model). It was acknowledged that there is some need
to define krill `management regions' in place of more pro rata designated Statistical
Areas/Subareas (CCAMLR, 1992). The demarcation of such regions was recognised as needing to account for equivocal evidence on the potential effects of
localised krill fishing on land-based predators (Agnew, 1993; Sushin & Myskov,
1993; Kerry et al., 1993)(Section 12.6, Overlap of krill fishery and predators). Following work in 1992, CM 46/Xl (Table 12.4) was adopted as an interim allocation of
catches by Statistical Subarea within Area 48 for the 1992±1993 and 1993±1994
seasons (CCAMLR, 1992). However, this measure subsequently lapsed and its
replacement awaits the results of attempts to refine alternative subarea allocations
based on krill, fishery, predator and environmental considerations (e.g. along the
lines outlined by Watters & Hewitt, 1993 and SC-CAMLR, 1999).
The second, far-reaching development was the elaboration of a decision rule to
address the requirements of predators (CCAMLR, 1994). In early application of the
krill yield model, allowance for predators had been discussed and a `discount factor'
to proportionately reduce calculated yield was proposed. The difficulty with this
suggestion, of course, was that there was no way of determining an appropriate level
for the discount factor, and in any case it was argued that krill-predator needs were
taken into account implicitly through estimation of krill M. As an alternative, the
population projections of the model were used to define a second decision rule to
complement the rule safeguarding against critical reductions in stock biomass
(Table 12.3). Together, both decision rules act in a conservative manner to ensure
that krill spawning stock as well as predator (ecosystem) needs are safeguarded. By
adopting such rules, CCAMLR finally achieved an explicit formulation of the
operational criteria (Section 12.3, Formulating operational management objectives)
it had originally identified as crucial for management of the krill fishery. Comparable rules have since been applied to species other than krill (e.g. Patagonian
Toothfish, Dissostichus spp.) (SC-CAMLR, 1998).
In 1994, CCAMLR accepted the revised yield model, final FIBEX Bo estimates,
and three-part decision rule, thereby completing the most important phase in the
development of a krill management procedure within four years (1991±1994). The
value of g now accepted for krill, following the three-part decision rule, is 0.116. Catch
limits for Divisions 58.4.2 and 58.4.1 have been set using this g and estimates of Bo
from Australian acoustic surveys (Pauly et al., 1996; see also Everson, Chapter 3).
However, although a revision of the FIBEX results in 1994 using new target
strengths for krill suggested that the precautionary yield in Area 48 should be 4.1
Krill Fisheries in the Southern Ocean
315
million tonnes, a new Conservation Measure has not yet been adopted. This is
attributable to:
(1)
(2)
a current lack of guidelines on which to base an allocation of catch limits to
Subareas within Area 48, and
a growing concern that the estimate of Bo from the 1981 FIBEX survey was
made in a year of high krill biomass that is no longer characteristic of the South
Atlantic.
Siegel et al. (1998) have demonstrated that, at least in the Antarctic Peninsula area,
krill biomass was an order of magnitude lower for most of the period 1985±1995 than
in 1981±1982. Although the biomass may have increased again in 1996 this is
thought to be temporary (SC-CAMLR, 1998 ± Annex 4, Appendix D, Fig. 12).
Therefore, as already mentioned, a synoptic survey of Area 48 was planned for early
2000 to establish a new Bo for krill in the Area. Adoption of any replacement for
CM 32/X also awaits the outcome of deliberations aimed at developing allocation
procedures for catch limits to subareas (SC-CAMLR, 1998). As the current catch is
perceived to be low (about 100 000 t and therefore less than 10% of the precautionary yield), there appears to be no immediate urgency to revise CM 32/X for
Area 48 (Table 12.4) until improved estimates of krill precautionary catch limits are
available based on, for example, the synoptic survey alluded to above.
Taking account of functional interaction(s) between krill and predators
Although the decision rules (Table 12.3) take explicit account of the needs of
predators, arbitrary levels of krill escapement have been set to meet such needs
commensurate with the rules. However, CCAMLR has realised that existing data
on predator performance and krill variability might facilitate explicit modelling of
the functional relationship between predator and krill populations. Consequently,
this may facilitate more objective definition of the levels of escapement necessary to
meet predator needs (Butterworth & Thomson, 1994, 1995).
The above approach initially utilised preliminary information on the population
dynamics of AdeÂlie penguins (Pygoscelis adelie), Antarctic fur seals (Arctocephalus
gazella), crabeater seals (Lobodon carcinophagus) and blackbrowed albatross
(Diomedea melanophrys) to identify functional relationships between juvenile and
adult predator survival rates with krill abundance (Butterworth & Thomson, 1995).
From the `one-way' interaction model developed (i.e. where fluctuations in krill
abundance impact on predator populations, but not vice versa), variability in annual
krill recruitment was shown to render predator populations less resilient to krill
harvesting than deterministic evaluations would suggest (Butterworth & Thomson,
1995). These initial results focused subsequent discussion on interpreting preliminary adult survival rate estimates for some of the predator populations used in
the model (SC-CAMLR, 1993).
316
Krill: Biology, Ecology and Fisheries
Further model developments (Butterworth & Thomson, 1994), introduced the
concept of krill `availability' to provide a random component in the relationship
between biomass and availability (i.e. with biomass as a function of spatial as well as
temporal variability). For the blackbrowed albatross and Antarctic fur seal in particular, an almost unrealistically poor resilience was observed for these two species
in relation to krill fishing, and by implication to changes in krill availability (even
without variability in krill recruitment being taken into account) (Butterworth &
Thomson, 1994). Currently, CCAMLR is now striving to ascertain whether the
modelling technique is inappropriate or whether there are inherent negative biases
in available estimates of predator survival rates from field data (SC-CAMLR, 1995,
1997).
In the absence of more quantitative assessments of predator responses to changes
in krill availability in general, and to different levels of krill escapement in particular, SC-CAMLR has accepted a target level of escapement of 0.75 as an initial
value on which to base management recommendations (SC-CAMLR, 1994). The
current krill escapement value of 0.75 lies between the customary 0.5 level applicable in the single species context and the 1.0 level commensurate with no fishing.
CCAMLR has acknowledged that this value may be revised as the modelling
initiatives described here are further refined.
Indirect approaches
Tracking indices in krill biomass
In its early years, and following realisation that age-based assessments of krill were
unlikely to be possible, CCAMLR considered using CPUE indices to monitor
trends in krill population biomass and in biomass dynamic models. In 1986, simulation studies were commissioned of the Soviet and Japanese krill fisheries
respectively to investigate the possibility of using CPUE as an index of krill biomass
(SC-CAMLR, 1986, 1987). These studies (Butterworth, 1989a, b; Mangel, 1989; SCCAMLR, 1988) concluded that certain catch-dependent indices, particularly those
containing some element of fishery search-time, could be used to assess krill
abundance. The CPUE simulation studies supported Shimadzu's (1985) and Everson's (1988a) earlier conclusions that catch-per-fishing time (i.e. CPH) provides the
most useful index of local krill abundance. They also indicated that various catch
and effort data might be utilised to derive a Composite Index of Krill Abundance
(SC-CAMLR, 1988).
A recent experiment to investigate the properties of a Composite Index found
that the frequency distributions of commercial catch-per-fishing time and krill
density from acoustic surveys showed similar forms, although non-random movement by a fishing vessel may obscure such comparisons (SC-CAMLR, 1994). A
further difficulty in using search-time is that fishing operations are limited by catch
processing efficiency rather than krill availability. Unfortunately, the conclusion of
Krill Fisheries in the Southern Ocean
317
this work is that CPUE is not a very good estimator of biomass and to date there
have been no further attempts to utilise it. Nevertheless, CCAMLR has continued
to encourage development of approaches to use CPUE data to monitor fishing
activities in relation to krill biomass. More specifically, it has endorsed the need to
collect information on fishing vessel activities at random times in an effort to
quantify search-time (SC-CAMLR, 1993).
Estimating krill potential yield from predator consumption
As early as 1990, CCAMLR discussed the use of predator consumption rates to
bound estimates of B0 and, consequently, krill potential yield (SC-CAMLR, 1990;
Agnew, 1992). Everson & de la Mare (1996) have recently refined these ideas, to
provide a method for estimating B0 from predator consumption rates around South
Georgia with:
Bo ˆ
PT…M2 ‡ V‰MŠ†
M3
…12:1†
where: P = annual krill consumption by land-based predators; M = annual krill
mortality rate; T = krill retention time within predator foraging area(s); V[M] =
variance of M estimate.
For example, Everson & de la Mare estimate that, at South Georgia, 9.76 million t
of krill are taken by predators annually, mostly within 100 km of the Island. With a
turnover rate of once a year, and estimates of M and V[M] of 0.6 and 0.1 respectively, the calculated instantaneous standing stock in the area of 9.5 million t is
considerably greater than the revised FIBEX estimate of 4.1 million t for Area 48 as
a whole (SC-CAMLR, 1994). This figure, when used with the estimate of g from the
potential yield model (0.116) infers an annual krill yield of 1.1 million t in the
vicinity of South Georgia.
12.5
Implications on krill management of links between krill
and environmental parameters
Krill flux
CCAMLR has recognised that application of both direct and indirect approaches is
crucially dependent on movement (flux) of krill between various localities. Subsequently, developments in 1993 focused on the effect of krill flux (either passively as
a result of water movement or actively through migration) in estimating yield,
primarily as a consequence of potential impacts on the value of B0 (SC-CAMLR,
1993).
A CCAMLR Workshop on Krill Flux in 1994 used data from satellite-tracked
drifters to investigate passive transport of krill by prevailing easterly currents
318
Krill: Biology, Ecology and Fisheries
through the Scotia Arc (see Everson, Chapter 3). Results indicated water transport
times between the Antarctic Peninsula and South Georgia of about six months (Ichii
& Naganobu, 1996) and, as originally postulated by Marr (1962), the effects of this
circulatory system may link krill populations around the South Shetlands, South
Orkneys and South Georgia. Despite this work and general support for its conclusions from the results of the CCAMLR Workshop on Area 48 (SC-CAMLR,
1998), the extent to which krill are exchanged between areas (through passive
transport around the Antarctic Continent) or remain resident in highly productive
areas (i.e. the South Shetland Islands) remains unknown.
The consequences of uncertainty about flux hold significance for management,
especially in the setting of precautionary limits for subareas within Area 48 ±
effectively the `management areas' used by CCAMLR (see Section 12.3). Should
flux be important, the available krill stock in any one area will be greater than that
estimated by a `snapshot' survey. If krill are essentially resident in one area, then
stock size will be equivalent to that seen in a survey; assuming that there is no flux
results in a conservative yield estimate (SC-CAMLR, 1994), and that has been the
approach adopted by CCAMLR to date. However, this approach does assume that,
should flux be important, there is no fishing in upstream areas. Obviously, heavy
fishing in upstream areas could remove all krill from downstream areas if flux was
important, but would have no effect if the contrary was true.
In contiguous areas, such as Subareas 48.1, 48.2 and 48.3, flux-modified catch
limits might necessitate further division of catch between subsidiary areas within an
overall areal catch in order to take into account possible fluxes (i.e. movement) of
krill biomass into such areas from a single upstream source (i.e. Subarea 48.1). Thus,
the total yield estimate for Area 48 would remain the same, but the catch limits for
individual subareas would change. Balanced against the current approach that not
taking flux into account remains conservative, CCAMLR is increasingly aware that
in the case where krill flux rates are not constant between areas, a simple pro rata
allocation of some total precautionary catch for Area 48 to Subareas could lead to
inappropriately high local catches in some smaller areas (SC-CAMLR, 1992
onwards).
Sea-ice
Apart from water circulation, the environmental factor most likely to influence krill
distribution and abundance is sea-ice (Mackintosh, 1972, 1973). A recent examination of long-term data on ice concentration has established that although total
Antarctic sea-ice cover may remain fairly constant from year to year, anomalies in
the northern ice extent progress clockwise around the Continent with a periodicity
of approximately 7±9 years (Murphy et al., 1995; Stammerjohn & Smith, 1997); a
situation reflected further in local sea-ice distribution.
Variability in krill recruitment is thought to be linked to such cycles in the
Antarctic Peninsula region. Kawaguchi & Satake (1994), Siegel & Loeb (1995) and
Krill Fisheries in the Southern Ocean
319
Loeb et al. (1997) have reported that winters with a substantial extent of sea-ice are
followed by summers of high krill recruitment, as shown by high proportions of
juvenile krill (i.e. animals spawned the previous summer) taken in both commercial
and research net hauls. Such observations may be attributable to winter sea-ice
offering an important feeding/nursery ground and refugium for both adult and
larval/juvenile krill (following suggestions by Hamner et al., 1989; Daly, 1990).
These results also intimate that krill recruitment is largely independent of spawning
stock size in the region. However, Siegel & Loeb (1995) suggest that prolonged
winter sea-ice cover leads to early krill maturation and spawning with subsequent
enhanced recruitment to the next season's ice. Consecutive years of heavy sea-ice
cover therefore act in concert to increase krill recruitment (Loeb et al., 1997) while
the absence of sea-ice may influence the dominance of either krill or salps at different times (Loeb et al., 1997).
Possible associations between sea-ice and krill CPUE are investigated by comparing ice cover in October of one split-year with catch-per-hour (CPH) in that splityear (see Section 12.2 for description of analysis). In Subarea 48.1, a strong positive
correlation is observed between CPH during the period January to March, when
fishing usually takes place (see Section 12.2), and sea-ice conditions the previous
winter (Table 12.5). This may be due to such effects as noted in the previous
paragraph, but it is not possible to confirm whether the observed increase in CPH in
Subarea 48.1 is due to increased density of adult krill or the occurrence of large
numbers of recruiting krill. Nevertheless, and irrespective of how it is mediated, any
correlation between krill CPH and ice may have operational consequences for the
fishery.
By contrast, negative or weak associations are evident between CPH and sea-ice
in Subareas 48.2 and 48.3 with Fedulov et al. (1996) finding a correlation between
Table 12.5 Correlation coefficients comparing CPH (catch/hours fished) and sea-ice cover in
October. For description of analysis refer to Section 12.2, Available data, of text. CHL = Chile, JPN
= Japan, POL = Poland and SUN = Soviet Union (* ± Significant at 0.05 level and ** ± significant at
0.01 level).
Subarea
Nation
Ice concentration
n
Subarea
48.1
48.1
48.2
48.2
48.3
48.3
@
JPN
CHL
JPN@
SUN+
SUN
POL
88.3.1#
48.1
48.2 & 48.3
0.36
0.10
70.66*
70.21
70.52
70.10
0.75*
0.58
70.54
70.51
0.24
0.30
0.87**
0.78**
70.22
70.06
0.44
0.00
Excluding 1990 when only 1 tonne of krill caught
Bounded by 758S, 708W; 608S, 908W
+
Combined Russian & Ukrainian Fisheries Data from 1991/92
#
14
10
12
9
9
9
320
Krill: Biology, Ecology and Fisheries
krill catch-per-day fished and sea-ice for the Russian fishery around South Georgia
during the period June±August. One implication of this finding is that sea-ice
conditions may influence CPUE throughout Area 48. Although the data available in
the CCAMLR database do not allow a detailed breakdown of CPH for the Russian
fishery, some positive correlation is evident between CPH in the Soviet fishery and
ice concentration the previous October (Table 12.5).
The cause of negative correlations between sea-ice and krill CPH in Subarea 48.2
shown in Table 12.5 remains unclear. Agnew (1993) has previously reported that the
fishery in Subarea 48.2 is erratic, perhaps as a result of a variable current structure
around the South Orkneys. It may be that years of heavy sea-ice concentration are
associated with highly disrupted current flow fields around the Islands thereby
resulting in different krill retention times in the region and hence variable catch
rates.
From these results, linking sea-ice and the performance of the krill fishery in the
forthcoming year offers an intriguing possibility for predicting catch rates based on a
relatively simple environmental index (i.e. sea-ice cover). The matter is obviously
one that warrants serious and further consideration.
12.6
Monitoring the effects of krill harvesting
Predator indices
As long ago as 1981, Everson (1981) indicated the likely effects on krill-dependent
predators of intensive krill fishing in restricted area(s) (Table 12.6); a concern
specifically addressed by CCAMLR Article II, Sub-paragraph (3).(b) (Anon.,
1995). Coupled with possible effects of fishing on krill availability to predators, it is
assumed that it will be possible to detect changes in krill availability through some
index of predator performance (Croxall, 1990). Predator data would therefore need
to be incorporated into the management paradigm (see also Section 12.3). Apart
from the points already discussed in Section 12.4, CCAMLR has chosen to address
this problem in various, but not mutually exclusive, ways. CCAMLR's first step was
to develop an Ecosystem Monitoring Programme (CEMP)(SC-CAMLR, 1984,
1985) aiming:
`To detect and record significant changes in critical components of the ecosystem
to serve as the basis for the conservation of Antarctic marine living resources. The
monitoring system should be designed to distinguish between changes due to the
harvesting of commercial species and changes due to environmental variability,
both physical and biological.'
A system to regularly record selected life-history parameters of key seabird and
seal populations has been in place since 1986 (SC-CAMLR, 1986). The programme
involves using standard procedures to monitor a number of parameters of selected
Krill Fisheries in the Southern Ocean
321
Table 12.6 Postulated effects of intensive krill fishing on predators in restricted area(s) (after
Everson, 1981 and Miller, 1991).
Impacted stocks
Single krill stock
Several areal krill stocks
Species close to/in fishing area (e.g. breeding seabirds)
(i) Heavy fishing prior to
critical predation period
each season
Significant in fishing area,
elsewhere minimal
Significant: Increasing with
time to level-off eventually
(ii) Intensive fishing during
and after critical predation
period
Small
Significant in subsequent
years
Species not tied to fishing area (e.g. whales)
(i) Intensive fishing before
predator normally present in
fishing area
Small overall
Reduced density (i.e. feeding
elsewhere)
(ii) Intensive fishing during
and after predator present in
fishing area
None or slight reduction in
density
Slight short-term reduction
Major long-term reduction
predator species (generally relating to breeding and foraging success), parameters
estimating krill abundance and environmental parameters such as sea-ice extent
(Anon., 1997, see also Everson, Chapter 7). Since 1992 SC-CAMLR has undertaken
annual `ecosystem assessments' using the data collected by CEMP. Agnew & Nicol
(1997) have provided a comprehensive review of the Programme and research into
multivariate analyses of CEMP parameters, especially, in detecting `unusual' years
(of krill availability or otherwise), continues to be a high CCAMLR priority (SCCAMLR, 1997, 1998).
Overlap of krill fishery and predators
Despite calculations of krill yield that now take into account krill and predator
requirements, CCAMLR has been acutely aware of the potential for local competition between predators and the krill fishery. From a global perspective (e.g.
within a Subarea or Area), fishing mortality (F) might remain within the limits set
by the yield model and so provide sufficient escapement for predator needs.
However, at a more localised scale, F might be much greater and krill escapement
therefore too low to sustain predators with restricted foraging ranges. This concern
is exacerbated by the knowledge that the foraging of many bird and seal predators is
restricted during breeding to close to land for about four months in the austral
summer (Croxall, 1990; Croxall et al., 1984, 1992).
CCAMLR's debate on possible competition between the krill fishery and krill
predators has been largely influenced by an improvement in the availability of finescale, fisheries information for Subarea 48.1. From Fig. 12.3, it is apparent that
322
Krill: Biology, Ecology and Fisheries
although only about 16% of the total krill catch has come from Subarea 48.1, the
annual catch has been relatively consistent at about 40 000 to 80 000 t. Based on an
earlier description of haul-by-haul data from the Chilean fishery (Marin et al., 1992),
Agnew (1993) has shown that this catch is highly concentrated on the shelf areas of
the South Shetland Islands, mostly between December and March, with patterns of
high catch being largely predictable from year to year. The areas concerned support
some of the largest concentrations of seabirds, including penguins, in the CCAMLR
Area as well as substantial seal and whale populations (Everson, 1977; Laws, 1989).
Agnew & Nicol (1997) have emphasised that krill's important ecological status
and contagious distribution result in a variety of possible interactions with other
biota as well as with the environment over a wide range of spatial/temporal scales
(see also Murphy et al., 1988; Murphy, 1995). It is the functional relationship
between predators, prey and the fishery at each scale which is crucial to determining
whether potential overlap between fishing and predator foraging is important.
Progress has been made in modelling functional relationships between krill and
predators (Mangel & Switzer, 1998) following earlier work by Mangel (1994).
At the local scale (< tens of kilometres), Ichii et al. (1994a) have indicated that the
fishery usually targets aggregation centres while penguins and the fishery may
exploit the same krill aggregations around the South Shetland Islands. On the other
hand, Bengtson et al. (1993) have shown that Chinstrap penguins (Pygoscelis antarctica) feed in the upper levels of such aggregations. Ichii et al. go on to suggest that
the fishery and penguins exploit different krill size classes.
At a slightly larger scale (tens to fifties of kilometres), Agnew (1993) found
patterns of fishing from 1988 to 1992 within Subarea 48.1 to be consistent between
years. Seventy to 90% of the total catch for the Subarea was taken between January
and March, within 100 km of shore-based penguin breeding colonies each year. The
period December to March is the peak breeding time, when penguins are restricted
to foraging distances within 100 km of their breeding colonies (Croxall et al., 1985).
Agnew's (1993) results prompted Ichii et al. (1994b) to consider the spatial distribution of penguin colonies and krill catches in more detail. They showed that
even though a high percentage of the krill catch in Subarea 48.1 may be taken within
the foraging ranges of land-based penguin colonies, the largest of these colonies are
some distance from the fishing grounds. More specifically, the Japanese krill fishery
is generally concentrated close to Livingston and Elephant Islands, while the largest
penguin colonies are located on Low, Nelson and King George Islands (WoÈehler,
1993).
The conclusions of Ichii et al. (1994b) are largely supported by studies showing
that the foraging ranges of Chinstrap and AdeÂlie penguins are also generally confined to within 100 km of their breeding colonies (Lishman, 1985; Everson, 1981,
1988b; Trivelpiece et al., 1987). However, Kerry et al. (1993) have shown that AdeÂlie
penguins on Bechervaise Island (within Subarea 58.4) forage several hundreds of
kilometres from land to feed on krill aggregations over the continental shelf break
(i.e. where krill fishing took place in the early 1980s). If AdeÂlie penguins are also
Krill Fisheries in the Southern Ocean
323
able to exhibit similar foraging ranges in the Antarctic Peninsula (Subarea 48.1)
area then some overlap with the fishery is still possible.
A further difficulty in interpreting the effects of potential overlaps between krill
fisheries and predator foraging areas may be related to the krill flux issue. If
important, this could lead to significant variability in the availability of krill to
predators. For instance, based on modelling studies of krill availability, predator
demands and water flow around South Georgia, Murphy (1995) has shown the
importance of the magnitude and timing of horizontal fluxes in determining krill
abundance available to predators. The situation at the Antarctic Peninsula is not so
clear. While there may be longshore movements, and a krill flux in and out of
various locations, there is also now considerable evidence to suggest that some krill
stocks may remain resident in specific areas and that adult animals migrate on to the
shelf to spawn (Ichii & Naganobu, 1996). The likely impact of overlap between
fishing and predators as a function of different levels of flux and residency has yet to
be adequately investigated. However, Mangel & Switzer (1998) have examined such
a possibility in a model they have recently developed.
An area which remains unexplored pertains to the extent to which overall predation on krill can be attributed to land-based predators. There are a number of
marine predators (fish, squid and whales) which have not yet been taken into
account in fishery, krill and krill predator overlap studies. At least in the past, this
predation has been substantial (Everson, 1977; Miller & Hampton, 1989).
Irrespective of the ultimate interaction between predators and fishery, CCAMLR
has developed a suite of calculations to track the extent of the potential overlap
between the two. Agnew & Phegan (1995) developed a model of land-based
predator foraging demands at a small spatial scale. This model was used to develop
an overlap index which expressed krill catch relative to predator foraging demands.
The results (Fig. 12.6) show a progressive decrease in the index in Subarea 48.1
during the critical period when predators are tied to land-based breeding sites. The
decrease is most apparent from 1991 onwards, the period during which CCAMLR
has focused its attention increasingly on the fishery-landbased predator overlap
problem at critical times for the latter (SC-CAMLR, 1991±1994). As shown in
Section 12.2, the above can only partially be explained by increasing the availability
of ice-free areas later in the season (Fig 12.5). Recently, CCAMLR has developed a
variety of other overlap indices for predators and fishery, including the SchroÈeder
index which reflects relative degrees of overlap (SC-CAMLR, 1998, 1999; Everson,
Chapter 8).
At current levels of fishing, in Subarea 48.1 at least, krill escapement from the
fishery is probably adequate to meet predator demands. Chinstrap penguins constitute by far the most numerous krill predator in the Subarea, and the most significant in terms of krill consumption in most of Area 48 (Croxall et al., 1984, 1985).
One and a half million Chinstrap pairs populate the South Shetland Islands, compared with 60 000 AdeÂlie and 34 000 Gentoo penguins (Pygoscelis papua); over two
million pairs of all three species are located in Subarea 48.1 (WoÈehler, 1993). Total
0
1984
20000
40000
60000
80000
100 000
120 000
1985
1986
1987
1988
1989
1991
Split year
1990
1992
1993
1994
1995
CPD
TCS
1996
1997
Fig. 12.6 Krill catch in tonnes by split year taken within the critical distances of various land-based predator colonies within
Subarea 48.1 during the breeding period (CPD) compared with the total catch in the Subarea (TCS).
Tonnes
324
Krill: Biology, Ecology and Fisheries
Krill Fisheries in the Southern Ocean
325
krill consumed by these species has been calculated as 370 000 t for the South
Shetland populations (Agnew, 1993) during the period (December to March) when
foraging is restricted close to breeding sites. With the current krill catch being of the
order of 40 000 to 80 000 t annually, and the krill biomass in Subarea 48.1 over the
austral summer probably being greater than two million tonnes (Siegel, 1992;
Trathan et al., 1993, 1995), there is little suggestion that the latter is unable to
support both penguins and other krill predators such as fish. However, should the
catch in Subarea 48.1 increase substantially, the issue of overlap of predators and the
fishery will once again assume prominence.
Effects of management measures on fishing
Concomitant with developing precautionary catch limits for krill, CCAMLR has
identified a number of other approaches (e.g. reactive management, predictive
management, open/closed areas, indirect methods, pulse fishing and feedback
management) which may serve to manage krill fishing (SC-CAMLR, 1991±1993).
The scope of these approaches and the resulting dialogue with fisherman have
served to emphasise the need to evaluate objectively the effects of any management
measures which might be introduced. CCAMLR has therefore accepted (SCCAMLR, 1993) that while adequate protection should be afforded to krill-dependent predators at critical times and in specific areas, such protection should not
exert unnecessary or unreasonable restrictions on the fishery. Various models have
therefore been developed to assess the potential effects of selected management
measures on the fishery.
Agnew & Marin (1994) constructed a spatial model based on the Chilean krill
fishery to simulate the behaviour of the fishery in Subarea 48.1. The model takes
into account fishing duration, fishing vessel performance, CPH and prevailing
management measures in force. Agnew (1994) subsequently extended the model to
include Japanese data, an estimate of encounter probability with fishable aggregations and to account for variability in CPUE as well as other parameters. In
addition, the model was applied to fleets rather than single vessels and the performance of management measures was assessed by effects on:
(1)
(2)
the total catch of the fleet over the whole fishing season, and
potential interaction(s) between catches and foraging predators at specific
times (i.e. during critical periods when predators are breeding).
Out of a range of management measures, Agnew (1994) found that a two-month
closure (January to February) offered the most successful strategy for both the
Chilean and Japanese fleets within 75 km of selected predator areas. This impacted
least on the fishery, although some small overlap between predators and fishery
remained.
The continued development of management models (SC-CAMLR, 1997) has
exciting strategic implications since it provides a means to evaluate the potential
326
Krill: Biology, Ecology and Fisheries
effects of specific management action on both predators and fishery. Such models
also serve to identify key parameters which may be used to test management procedures and to advance consideration of how specific indices of predator status, or
performance, may be taken into account in management measure formulation (de la
Mare, 1986).
12.7
Discussion
The emergence of many Antarctic fisheries in the early 1970s was undoubtedly
predicated by developments at the Third United Nations Conference on the Law of
The Sea (UNCLOS) (Boczek, 1983). As a consequence, control of most of the
world's fisheries was gradually transferred from international arrangements to
Coastal State jurisdiction within established national, 200-mile Exclusive Economic
Zones (EEZs) (May et al., 1979). However, unresolved sovereignty issues in the
Antarctic have generally precluded the establishment of universally accepted EEZs
(Freestone, 1995).
Antarctic fisheries have therefore been generally `open to access' and have had
much in common with the `boom and bust' scenarios demonstrated by Antarctic
whaling. Under such circumstances, and in the absence of effective international
control, Antarctic fisheries have tended to move from one depleted species to the
next (Kock, 1992). This pattern is clearly illustrated by trends in finfish catches
following severe depletion of Antarctic cod (Notothenia rossii) stocks post-1972 (Fig.
12.7). With an overall steep decline in catches of finfish, a shift of focus to serious
development of the krill fishery thus became inevitable from about 1978 onwards.
However, it should be noted that the mackerel icefish (Champsocephalus gunnari) fishery was a notable exception from the general boom or bust scenario
associated with Antarctic fisheries. From 1978, catches of this species demonstrated
a clear five-yearly cycle (Fig. 12.7). The first two of these (1978 and 1983) were
followed by significant increases in krill catches within one or two years. In the latter
case, the increase in icefish catches between 1982 and 1984 was associated with a
marked reduction in krill catches. Independently, there was also a marked decline in
krill catches just prior to, and immediately following, the break up of the Soviet
Union in 1991. Together, these examples clearly suggest, like the finfish fishery,
operational considerations (i.e. selective access to resources) have exerted a profound influence on krill fishery development.
As emphasised, considerable concern came to be expressed about possible overexploitation of krill almost as soon as the fishery began to flourish. Proponents of
moderation warned that given krill's perceived high abundance, the potential for
rapid and uncontrolled growth of the fishery was high following precedent set by the
explosive, and essentially unsustainable, expansion in the fisheries for a number of
finfish (Fig. 12.7) in combination with the demise of the Peruvian anchovy harvest
between 1958 and 1970 (Reid, 1988).
0
100 000
200 000
300 000
400 000
500 000
1983/84
1982/83
1981/82
Split year
KRI
ANI
TOP
NOR
1994/95
600 000
NOS
1996/97
1993/94
1992/93
1991/92
1990/91
1989/90
1988/89
1987/88
1986/87
1985/86
1980/81
1979/80
1978/79
1977/78
1976/77
1975/76
1974/75
1973/74
1972/73
1971/72
1970/71
1969/70
Fig. 12.7 Annual catches of major species in the CAMLR Convention Area from 1969/70 to 1996/97. ANI = Champsocephalus gunnari
(Mackerel icefish); KRI = Euphausia superba (Krill); NOR = Notothenia rossii (Marbled rockcod); NOS = Notothenia squamifrons (Grey
rockcod); NOT = Patagonotothen guntheri (Patagonian rockcod) and TOP = Dissostichus eleginoides (Patagonian toothfish).
Tonnes
NOT
1995/96
700 000
Krill Fisheries in the Southern Ocean
327
328
Krill: Biology, Ecology and Fisheries
The emergence of CCAMLR was the response to these concerns. Despite its
notable successes in formulating a scientifically defensible, management policy for
the krill fishery (Section 12.3), various factors have retarded the CCAMLR's efforts
to address issues associated with krill management (Nicol, 1991). Croxall et al.
(1992) have emphasised that CCAMLR's relatively slow progress was due to conceptual as much as practical difficulties. These necessitated a new approach to
managing the fishery and were compounded by krill's contagious distribution, the
fishery's potentially large geographical range and the assertions of a number of krillfishing nations that:
. krill catches are small in relation to the available stock;
. there is no intention to increase krill catches dramatically in the near future;
. krill management must be based on the `best scientific information' available
and hence on scientifically formulated assessments. Such assessments are currently limited as data are inadequate, particularly on:
(a) krill abundance, distribution and flux;
(b) functional relationships between krill and predators, and
. historical catch levels do not offer a scientific basis for managing the krill fishery.
With an improved flow of data from the fishery, these restrictions have been largely
circumvented by CCAMLR's acceptance of guiding principles, recognising SCCAMLR's scientific advice as being the `best available', adopting a precautionary
approach to management in the face of uncertainty and exercising feedback rather
than reactive management control (Section 12.3).
A second major factor which has undoubtedly influenced progress on management issues has been CCAMLR's requirement that decisions on matters of substance (i.e. conservation measures) be taken by consensus [Convention Article
XII.(1)] (Croxall et al., 1992). While this may have slowed down the making of
decisions, it has ensured that all parties to the Convention are included in the
decision-making process and that, in the absence of objections, decisions enter into
force within a period of six months. Since CCAMLR, like so many international
organisations, has no legal personality of its own, members are morally and strongly
obligated to enforce management measures, especially when these have been
agreed through a consensus approach.
Despite the above limitations, CCAMLR's progress in establishing formal conservation measures (including precautionary catch limits for krill) has compared
favourably with other fisheries commissions and conventions elsewhere in the world
(Miller, 1991) (Table 12.7). Indeed, CCAMLR has adopted various landmark
measures (e.g. first measure, size limit, catch limit, closed areas etc.) more rapidly
than most other international fisheries arrangements. Furthermore, CCAMLR
remains the only international fisheries commission to date to have formally
adopted any precautionary limit on catch in the interests of minimising the risk of
over-exploitation and to reduce possible ecosystem effects.
Krill Fisheries in the Southern Ocean
329
Table 12.7 Development and enforcement of Conservation Measures by CCAMLR compared
with the International Commission for the Southeast Atlantic Fisheries (ICSEAF, 1984) and the
International Commission for the Conservation of Atlantic Tunas (ICCAT, 1985). Values in
parentheses represent the time elapsed since the specific convention came into force while those in
bold pertain directly to the krill fishery.
Attribute/Measure
Number of members
Date of entry into force
First management measure
Fish size limit
Target fishing mortality
Total allowable catch (quota)
Mesh size regulation
Close season(s)
Closed area(s)
Catch reporting system
Inspection scheme
Observation scheme
Precautionary catch measure
Catch trigger/subareal catch allocation
Notification of new fishery
Ecosystem Protection
Prohibition of Fishing Without a Licence
Satellite Vessel Monitoring
Trade Documentation
CCAMLR
ICSEAF
ICCAT
22
1982
1984 (2) (10)
1994 (13)
1987 (5)
1987 (5)
1984 (2)
1987 (5)
1984 (2)
1987 (6) (8)
1987 (6)
1992 (11)
1991 (10)
1991/92 (10/11)
1991 (10)
1991 (10)
1998 (16)
1998/2000 (16)
1999 (17)
17
1971
1973 (2)
±
?
1976 (5)
1973 (2)
1977 (6)
1979 (8)
1982 (11)
1976 (5)
?
±
±
±
±
±
±
±
22
1969
1973 (3)
1974 (5)
1975 (6)
1982 (13)
±
1994 (25)
1982 (13)
?
1974 (6)
?
±
±
±
±
±
±
?
Notwithstanding CCAMLR's obvious successes, uncertainties still surround the
future of the Antarctic krill fishery. The first concerns the fishery's possible course
of development. Obviously, the initial anticipated rapid expansion of the fishery has
not occurred and it seems that with the break up of the heavily subsidised Soviet
fishery the potential for such expansion has probably diminished. Such uncertainty
is further compounded by a lack of complete economic analyses of the fishery to
date.
Although economic analyses (e.g. Grantham, 1977; McElroy, 1980; Budzinski et
al., 1985) have focused on technological problems associated with krill processing
and product quality, thorough analyses of krill's potential market demand remain
limited. Both Anon. (1978) and Budzinski et al. (1985) have emphasised that
determining the prices of, and demand for, krill are the most persistent problems in
forecasting the fishery's future development. In this context, CCAMLR's on-going
call for industrial information from krill fishing nations (e.g. SC-CAMLR, 1991),
along with the precautionary catch limits in place, may be considered as a tactic to
improve knowledge of the fishery's economic driving forces while ensuring that the
risk of uncontrolled expansion is minimised. Even so, there is little information to
explain the economic motives for the observed increase in krill catches by Ukraine
during 1994±1995 compared with the 1993±1994 season, or for the stated intention
to cease krill fishing by Russia and Chile from 1995 onwards (SC-CAMLR, 1995).
330
Krill: Biology, Ecology and Fisheries
These examples serve to stress the urgency for economic studies of the fishery's
potential in furthering the development of effective management measures.
The second, and in terms of the convention's objectives probably more relevant,
uncertainty is to what extent, as well as how, ecosystem (i.e. multi-species) considerations may be more formally subsumed into management measures. While
some areas where progress has been made (Section 12.3) have been highlighted, the
topic is one which obviously warrants priority attention and development in the
future.
Finally, to address these various uncertainties still requires more information (e.g.
on search-time by fishing vessels) than is currently forthcoming from the krill
fishery. Only with such information will it be possible to assess objectively the
potential utility of regularly monitoring selected indices or fishery's factors (e.g.
length frequency composition of catches) crucial to developing realistic management procedures. The placement of suitably qualified observers aboard krill fishing
vessels would go a long way to improving the flow of data necessary to effectively
monitor the fishery (e.g. SC-CAMLR, 1990; CCAMLR, 1991) and to assess its
impact on available stocks as well as to evaluating future management action.
CCAMLR initiatives in this regard (e.g. Ichii et al. 1993; 1994a, b; Mujica et al., 1993;
Vagin et al., 1993) are therefore to be encouraged and their results are awaited with
interest.
Acknowledgements
We thank our many CCAMLR colleagues who have given their time, wisdom and
data to developing a sensible krill management regime. In particular, we greatly
appreciate useful comments from, and discussions with, Professors D.S. Butterworth and J. Croxall, Drs I. Everson, T. Ichii, K-H. Kock, V. Marin, E. Murphy, S.
Nicol and W. de la Mare. We dedicate this chapter to the memory of Dr Rodion
Makarov who passed away so tragically on 12 August 1994.
References
Agnew, D.J. (1992) Attempts at a basic accounting for Subareas 48.1 and 48.2. SC-CAMLRXI, Annex 4, Appendix F. CCAMLR, Hobart. pp. 174±76.
Agnew, D.J. (1993) Distribution of krill (Euphausia superba) catches in the South Shetlands
and South Orkneys. In Selected Papers Presented to the Scientific Committee of CCAMLR,
1992. CCAMLR, Hobart. pp. 287±303.
Agnew, D.J. (1994) Further development of a krill fishery simulation model. Document WGJoint-94/4. CCAMLR, Hobart. 10 pp.
Agnew, D.J. & Marin, V.H. (1994) Preliminary model of krill fishery behaviour in Subarea
48.1. CCAMLR Sci. 1, 71±79.
Agnew, D.J. & Nicol, S. (1997) Marine disturbances ± Commercial fishing. In Foundations for
Krill Fisheries in the Southern Ocean
331
Ecological Research West of the Antarctic Peninsula (R.M. Ross, E.E. Hofmann & L.B.
Quetin, eds), pp. 417±35. Antarctic Research Series No. 70. American Geophysical Union,
Washington.
Agnew, D.J. & Phegan, G. (1995) Development of a fine-scale model of land-based predator
foraging demands in the Antarctic. CCAMLR Sci. 2, 99±110.
Anon. (1978) The Antarctic krill resource: Prospects for commercial exploitation. Tetra
Technical Report. TC-903. 149 pp.
Anon. (1986) Post-FIBEX acoustic workshop. Frankfurt, Federal Republic of Germany,
September 1984. BIOMASS Report Series 40, 106 pp.
Anon. (1995) Commission for the Conservation of Antarctic Marine Living Resources: Basic
Documents, 6th edn. CCAMLR, Hobart. 122 pp.
Anon. (1997) CEMP Standard Methods for Monitoring Studies 1991. CCAMLR, Hobart.
222 pp.
Beddington, J.R. & Cooke, J.G. (1983) The potential yield of fish stocks. FAO Fisheries
Technical Paper 242, 47 pp.
Bengtson, J.L. (1984) Review of Antarctic marine fauna. In Selected Papers Presented to the
Scientific Committee of CCAMLR, 1982±1984. Part I. CCAMLR, Hobart. pp. 1±126.
Bengtson, J.L., Croll, D.A. & Goebel, M.E. (1993) Diving behaviour of chinstrap penguins at
Seal Island. Antarctic Sci. 5, 9±15.
Boczek, B.A. (1983) The protection of the Antarctic ecosystem: A study in international
environment law. Ocean Development and International Law Journal 13, 347±425.
Budzinski, E., Bykowski, P. & Dutkiewcz, D. (1985) Possibilities of processing and marketing
of products made from Antarctic krill. FAO Fisheries Technical Paper 268, 46 pp.
Burukovskiy, R.N. & Yaragov, B.A. (1967) Studying the Antarctic krill for the purpose of
organizing krill fisheries. In Soviet Fishery Research on Antarctic Krill (R.N. Burukovskiy,
ed.), pp. 5±17. U.S. Department of Commerce (TT 67-32683), Washington D.C.
Butterworth, D.S. (1986) Antarctic ecosystem management. Polar Rec. 23, 37±47.
Butterworth, D.S. (1989a) A simulation study of krill fishing by an individual Japanese
trawler. In Selected Papers Presented to the Scientific Committee of CCAMLR, 1988, Part I,
pp. 1±108. CCAMLR, Hobart.
Butterworth, D.S. (1989b) Some aspects of the relation between Antarctic krill abundance
and CPUE measures in the Japanese krill fishery. In Selected Papers Presented to the
Scientific Committee of CCAMLR, 1988, Part I, pp. 109±125. CCAMLR, Hobart.
Butterworth, D.S. (1990) Towards an initial operational management procedure for the krill
fishery in Subareas 48.1, 48.2 and 48.3. In Selected Papers Presented to the Scientific
Committee of CCAMLR, 1989, pp. 189±219. CCAMLR, Hobart.
Butterworth, D.S. & Thomson, R.B. (1994) Further calculations of the effects of krill fishing
on krill predators. WG-Krill-94/24. CCAMLR, Hobart. 25 pp.
Butterworth, D.S. & Thomson, R.B. (1995) Possible effects of different levels of krill fishing
on predators ± some initial modelling attempts. CCAMLR Sci. 2, 79±97.
Butterworth, D.S., Punt, A.E. & Basson, M. (1992) A simple approach for calculating the
potential yield from biomass survey results. In Selected Papers Presented to the Scientific
Committee of CCAMLR, 1991, pp. 207±215. CCAMLR, Hobart.
Butterworth, D.S., Gluckman, G.R., Thomson. R.B., Chalis, S., Hiramatsu, K. & Agnew, D.J.
(1994) Further computations of the consequences of setting the annual krill catch limit to a
fixed fraction of the estimate of krill biomass from a survey. CCAMLR Sci. 1, 81±106.
332
Krill: Biology, Ecology and Fisheries
CCAMLR (1990) Report of the Ninth Meeting of the Commission. CCAMLR, Hobart. 123 pp.
CCAMLR (1991) Report of the Tenth Meeting of the Commission. CCAMLR, Hobart. 101 pp.
CCAMLR (1992) Report of the Eleventh Meeting of the Commission. CCAMLR, Hobart.
99 pp.
CCAMLR (1994) Report of the Thirteenth Meeting of the Commission. CCAMLR, Hobart.
123 pp.
CCAMLR (1998) Report of the Seventeenth Meeting of the Commission. CCAMLR, Hobart.
166 pp.
Croxall, J.P. (1990) Uses of indices of predator status and performance in CCAMLR fishery
management. In Selected Papers Presented to the Scientific Committee of CCAMLR, 1989,
pp. 353±63. CCAMLR, Hobart.
Croxall, J.P., Ricketts, C. & Prince, P.A. (1984) The impact of seabirds on marine resources,
especially krill, at South Georgia. In Seabird Energetics (C.G. Whittow & H. Rahn, eds),
pp. 285±318. Plenum, New York.
Croxall, J.P., Ricketts, C. & Prince, P.A. (1985) Relationship between prey life-cycles and the
extent, nature and timing of seal and seabird breeding in the Scotia Sea. In Antarctic
Nutrient Cycles and Food Webs (P.R. Condy, W.R. Siegfried & R.M. Laws, eds), pp. 516±
33. Springer-Verlag, Berlin, Heidelberg.
Croxall, J.P., McCann, T.S., Prince, P.A. & Rothery, P. (1988) Variation in reproductive
performance of seabirds and seals at South Georgia, 1976±1986 and its implications for
Southern Ocean monitoring studies. In Antarctic Ocean and Resources Variability (D.
Sahrhage, ed.), pp. 261±85. Springer-Verlag, Berlin.
Croxall, J.P., Everson, I. & Miller, D.G.M. (1992) Management of the Antarctic krill fishery.
Polar Rec. 28, 64±66.
Daly, K.L. (1990) Overwintering development, growth, and feeding of larval Euphausia
superba in the Antarctic marginal ice zone. Limn. Oceanogr. 35, 1564±76.
de la Mare, W.K. (1986). Some principles for fisheries regulation from an ecosystem perspective. In Selected Papers Presented to the Scientific Committee of CCAMLR, 1986,
pp. 323±39. CCAMLR, Hobart.
de la Mare, W.K. (1994a) Modelling krill recruitment. CCAMLR Sci. 1, 49±54.
de la Mare, W.K. (1994b) Estimating krill variability and its recruitment. CCAMLR Sci. 1,
55±69.
Dolzhenkov, V.N, Lubimova, T.G., Makarov, R.R., Parfenovich, S.S. & Spiridonov, V.A.
(1989) Some specific features of the USSR krill fishery and possibilities of applying fishery
statistics to studies of krill biology and stocks. in Selected Papers Presented to the Scientific
Committee of CCAMLR, 1988, Part I, pp. 237±50. CCAMLR, Hobart.
Eddie, G.C. (1977) The harvesting of krill. Southern Ocean Fisheries Survey Programme.
GLO/SO/77/2. FAO, Rome. 76 pp.
Edwards, D.M. & Heap, J.A. (1981) Convention on the conservation of Antarctic marine
living resources: A commentary. Polar Rec. 20, 353±62.
El-Sayed, S.Z. (1988) Antarctic marine living resources: The BIOMASS Program. Oceanus.
31, 75±79.
El-Sayed, S.Z. & McWhinnie, M.A. (1979) Protein of the last frontier. Oceanus 22, 13±20.
Everson, I. (1977) The living resources of the Southern Ocean. Southern Ocean Fisheries
Survey Programme. GLO/SO/77/1. FAO, Rome. 156 pp.
Everson, I. (1978) Antarctic fisheries. Polar Rec. 19, 233±51.
Krill Fisheries in the Southern Ocean
333
Everson, I. (1981) Antarctic krill. BIOMASS Scientific Series 2, 31±45.
Everson, I. (1984) Marine interactions. In Antarctic Ecology, Vol. 2. (R.M. Laws, ed.),
pp. 783±819. Academic Press, London.
Everson, I. (1988a) Can we satisfactorily estimate krill abundance? In Antarctic Ocean and
Resources Variability (D. Sahrhage, ed.), pp. 199±208. Springer-Verlag, Berlin.
Everson, I. (1988b) Prey monitoring surveys. In Selected Papers Presented to the Scientific
Committee of CCAMLR, 1987, pp. 323±38. CCAMLR, Hobart.
Everson, I. & de la Mare, W. (1996) Some thoughts on precautionary measures for the krill
fishery. CCAMLR Sci. 3, 1±12.
Everson, I. & Goss, C. (1991) Krill fishing activity in the southwest Atlantic. Antarctic Sci. 3,
351±58.
Everson, I., Watkins, J.L., Bone, D.G. & Foote, K.G. (1990) Implications of new target
strength measurements for abundance estimates of Antarctic krill. Nature 345, 338±40.
FAO (1976) FAO Yearbook of Fishery Statistics, FAO, Rome. 336 pp.
Fedulov, P.P., Murphy, E.J. & Shulgovsky, K.E. (1996) Environment-krill relations in the
South Georgia marine ecosystem. CCAMLR Sci. 3, 13±30.
Foote, K.G., Everson, I., Watkins, J.L. & Bone, D.G. (1990) Target strength of Antarctic
krill (Euphausia superba) at 38 and 120 kHz. J. Acoust. Soc. America 87, 16±24.
Freestone, D. (1995) The conservation of marine ecosystems under international law. In
International Law and Conservation of Biodiversity (C. Redgwell & M. Bowman, eds),
pp. 91±107. Kluwer Law International, Great Britain.
Grantham, G.J. (1977) The utilization of krill. Southern Ocean Fisheries Survey Programme.
GLO/SO/77/3. FAO, Rome. 61 pp.
Greene, C.H., Stanton, T.K., Wiebe, P.H. & McClatchie, S. (1990) Acoustic estimates of
Antarctic krill. Nature 349, 110.
Gulland, J.A. (1977) Antarctic marine living resources. In Adaptations within Antarctic
Ecosystems (G.A. Llano, ed.), pp. 1135±44. Gulf Publishing, Houston.
Gulland, J.A. (1983) The development of fisheries and stock assessment of resources in the
Southern Ocean. Mem. Natl Inst. Polar Res., Special Issue 27, 129±52.
Hamner, W.M., Hamner, P.P. & Obst, B.S. (1989) Field observations on the ontogeny of
schooling of Euphausia superba furciliae and its relationship to ice in Antarctic waters.
Limn. Oceanogr. 34, 451±56.
Ichii, T. (1988) Observations of fishing operations on a krill trawler and distributional
behaviour of krill off Wilkes Land during the 1985/86 season. In Selected Papers Presented
to the Scientific Committee of CCAMLR, 1987, pp. 323±38. CCAMLR, Hobart.
Ichii, T. (1990) Distribution of Antarctic krill concentrations exploited by Japanese krill
trawlers and minke whales. In Proceedings of the National Institute of Polar Research
Symposium on Polar Biology 3, pp. 36±56.
Ichii, T. & Naganobu, M. (1996) Surface water circulation in krill fishing areas near the South
Shetland Islands. CCAMLR Sci. 3, 125±36.
Ichii, T., Ishii, H. & Naganobu, M. (1993) Abundance, size and maturity of krill (Euphausia
superba) in the krill fishing ground of Subarea 48.1 during the 1990/91 austral summer. In
Selected Papers Presented to the Scientific Committee of CCAMLR, 1992, pp. 183±99.
CCAMLR, Hobart.
Ichii, T., Naganobu, M. & Ogishima, T. (1994a) An assessment of the impact of the krill
fishery on penguins in the South Shetland Islands. CCAMLR Sci. 1, 107±13.
334
Krill: Biology, Ecology and Fisheries
Ichii, T., Naganobu, M. & Ogishima, T. (1994b) A revised assessment of the impact of krill
fishery on penguins in the south Shetland Islands. Document WG-Joint-94/17. CCAMLR,
Hobart. 20 pp.
ICCAT (1985) Basic Texts. International Commission for the Conservation of Atlantic
Tunas, Madrid. 99 pp.
ICSEAF (1984) Handbook of Regulatory Measures. International Commission for the
Southeast Atlantic Fisheries, Madrid. pp. 1±49.
Ivanov, B.G. (1970) On the biology of the Antarctic krill (Euphausia superba Dana). Mar.
Biol. 7, 340±51.
Kawaguchi, S. & Satake, M. (1994) Relationship between recruitment of the Antarctic krill
and the degree of ice cover near the South Shetland Islands. Fish. Sci. 60, 123±24.
Kerry, K., Clarke, J.R. & Else, G.D. (1993) The foraging range of AdeÂlie penguins at
BeÂchervaise Island, MacRobertson Land, Antarctica, and its overlap with the krill fishery.
In Selected Papers Presented to the Scientific Committee of CCAMLR, 1992, pp. 337±44.
CCAMLR, Hobart.
Knox, G.A. (1984) The key roÃle of krill in the ecosystem of the Southern Ocean with special
reference to the Convention on the Conservation of Antarctic Marine Living Resources.
Ocean Management 9, 113±56.
Kock, K.-H. (1991) The state of exploited fish stocks in the Southern Ocean: A review. Archiv
fuÈr Fischereiwissenschaft 41, 1±66.
Kock, K.-H. (1992) Antarctic Fish and Fisheries, 359 pp. Cambridge University Press, Cambridge.
Laws, R.M. (1977) Seals and whales of the Southern Ocean. Phil. Trans. Roy. Soc., London,
Series B. 279, 81±96.
Laws, R.M. (1985) The ecology of the Southern Ocean. Am. Scientist 43, 26±40.
Laws, R.M. (1989) Antarctica: The last frontier, 208 pp. Boxtree Publishers, Cambridge.
Lishman, G.S. (1985) The food and feeding of AdeÂlie and Chinstrap penguins at Signy Island,
South Orkney Islands. J. Zool. Soc. London 205, 245±63.
Loeb, V., Siegel, V., Holm-Hansen, O., Hewitt, R., Fraser, W., Trivelpiece, W. & Trivelpiece,
S. (1997) Effects of sea-ice extent and krill or salp dominance on the Antarctic food web.
Nature 387, 897±900.
Mackintosh, N.A. (1972) Life cycle of Antarctic krill in relation to ice and water conditions.
Discovery Rep. 36, 1±94.
Mackintosh, N.A. (1973) Distribution of post-larval krill in the Antarctic. Discovery Rep. 36,
95±156.
Mangel, M. (1989) Analysis and modelling of the Soviet Southern Ocean krill fleet. In
Selected Papers Presented to the Scientific Committee of CCAMLR, 1988, Part I, pp. 127±
235. CCAMLR, Hobart.
Mangel, M. (1994) Spatial patterning in resource exploitation and conservation. Phil. Trans.
Roy. Soc., London, Series B 343, 93±98.
Mangel, M. & Switzer, P.V. (1998) A model at the level of the foraging trip for the
indirect effects of krill (Euphausia superba) fisheries on krill predators. Ecol. Model 105,
235±56.
Marin, V.H., Mujica, A. & Eberhard, P. (1992) Chilean krill fishery: Analysis of the 1991
season. In Selected Papers Presented to the Scientific Committee of CCAMLR, 1991,
pp. 273±87. CCAMLR, Hobart.
Krill Fisheries in the Southern Ocean
335
Marr, J.W.S. (1962) The natural history and geography of the Antarctic krill (Euphausia
superba Dana). Discovery Rep. 32, 33±464.
May, R.M., Beddington, J.R., Clark, C.W., Holt, S.J. & Laws, R.M. (1979) Management of
multispecies fisheries. Science 205, 267±77.
McElroy, J.K. (1980) The economics of harvesting krill. In The Management of the Southern
Ocean (B. Mitchell and R. Sandbrook, eds), pp. 60±80. International Institute for
Environment and Development, London.
Miller, D.G.M. (1990) Commercial krill fisheries in the Antarctic, 1973±1988. In Selected
Papers Presented to the Scientific Committee of CCAMLR, 1989, pp. 229±81. CCAMLR,
Hobart.
Miller, D.G.M. (1991) Exploitation of Antarctic marine living resources: A brief history and a
possible approach to managing the krill fishery. S. Afr. J. Mar. Sci. 10, 321±39.
Miller, D.G.M. & Hampton, I. (1989) Biology and ecology of the Antarctic krill (Euphausia
superba Dana). BIOMASS Sci. Ser. 9, 166 pp.
Mitchell, B. & Sandbrook, R. (eds)(1980) The management of the Southern Ocean. In A
Report to the International Institute for Environmental Development, 162 pp. International
Institute for Environment and Development, London.
Mujica, A.R., AcunÄa, S. & Rivera, A.O. (1993) Krill population biology during the 1991
Chilean Antarctic krill fishery. In Selected Papers Presented to the Scientific Committee of
CCAMLR, 1992, pp. 223±35. CCAMLR, Hobart.
Murphy, E.J. (1995) Spatial structure of the Southern Ocean ecosystem: Predator-prey
linkages in Southern Ocean food webs. J. Anim. Ecol. 64, 33±347.
Murphy, E.J., Morris, D.J., Watkins, J.L. & Priddle, J. (1988) Scales of interaction between
Antarctic krill and the environment. In Antarctic Ocean and Resources Variability (D.
Sahrhage, ed.), pp. 120±30. Springer-Verlag, Berlin.
Murphy, E.J., Clarke, A., Simon, C. & Priddle, J. (1995) Temporal variation in Antarctic seaice: Analysis of long-term fast-ice record from the South Orkney Islands. Deep-Sea Res. 42,
1045±62.
Murphy, E.J., Constable, A. & Agnew, D.J. (1999) Modelling the dynamics of krill populations in the Antarctic Peninsula region. CCAMLR, Mimeo, WG-EMM-99/56. 18 pp.
Nemoto, T. & Nasu, K. (1975) Present status of exploitation and biology of krill in the
Antarctic. In Oceanology International 75. BPS Exhibitions, London. 353±60.
Nicol, S. (1989) Who's counting on krill? New Scientist 11, 38±41.
Nicol, S. (1991) CCAMLR and its approaches to management of the krill fishery. Polar Rec.
27, 229±36.
Nicol, S. & de la Mare, W.K. (1993) Ecosystem management and the Antarctic krill. Am.
Scientist 81, 36±47.
Nicol, S. & Endo, Y. (1997) Krill fisheries of the world. FAO Fish. Tech. Paper 367, 100 pp.
Pauly, T., Higginbottom, I., Nicol, S. & de la Mare, W. (1996) Results of a hydroacoustic
survey of Antarctic krill populations in CCAMLR Division 58.4.1 carried out in January to
April 1996. CCAMLR, Mimeo, WG-EMM-96/28. 22 pp.
Reid, W.V. (1988) Managing the Southern Ocean krill fishery. Resources. (Spring) 11±12.
Robinson, M.A. (1982) Prospects for world fisheries to 2000. FAO Fisheries Circular 722.
16.
SC-CAMLR (1984) Report of the Third Meeting of the Scientific Committee. CCAMLR,
Hobart. 238 pp.
336
Krill: Biology, Ecology and Fisheries
SC-CAMLR (1985) Report of the Fourth Meeting of the Scientific Committee. CCAMLR,
Hobart. 275 pp.
SC-CAMLR (1986) Report of the Fifth Meeting of the Scientific Committee. CCAMLR,
Hobart. 271 pp.
SC-CAMLR (1987) Report of the Sixth Meeting of the Scientific Committee. CCAMLR,
Hobart. 263 pp.
SC-CAMLR (1988) Report of the Seventh Meeting of the Scientific Committee. CCAMLR,
Hobart. 211 pp.
SC-CAMLR (1990) Report of the Ninth Meeting of the Scientific Committee. CCAMLR,
Hobart. 345 pp.
SC-CAMLR (1991) Report of the Tenth Meeting of the Scientific Committee. CCAMLR,
Hobart. 427 pp.
SC-CAMLR (1992) Report of the Eleventh Meeting of the Scientific Committee. CCAMLR,
Hobart. 487 pp.
SC-CAMLR (1993) Report of the Twelfth Meeting of the Scientific Committee. CCAMLR,
Hobart. 431 pp.
SC-CAMLR (1994) Report of the Thirteenth Meeting of the Scientific Committee. CCAMLR,
Hobart. 453 pp.
SC-CAMLR (1995) Report of the Fourteenth Meeting of the Scientific Committee. CCAMLR,
Hobart. 460 pp.
SC-CAMLR (1996) Report of the Fifteenth Meeting of the Scientific Committee. CCAMLR,
Hobart. 456 pp.
SC-CAMLR (1997) Report of the Sixteenth Meeting of the Scientific Committee. CCAMLR,
Hobart. 437 pp.
SC-CAMLR (1998) Report of the Seventeenth Meeting of the Scientific Committee.
CCAMLR, Hobart. 456 pp.
SC-CAMLR (1999) Report of the Eighteenth Meeting of the Scientific Committee. CCAMLR,
Hobart. 461 pp.
Shimadzu, Y. (1985) A brief summary of Japanese fishing activity relating to Antarctic krill,
1972/73 to 1982/83. In Selected Papers Presented to the Scientific Committee of CCAMLR,
1982±1984. Part I, pp. 439±71. CCAMLR, Hobart.
Siegel, V. (1992) Estimation of krill (Euphausia superba) mortality and production rate in the
Antarctic Peninsula region. In Selected Papers Presented to the Scientific Committee of
CCAMLR, 1991, pp. 159±76. CCAMLR, Hobart.
Siegel, V. & Loeb, V. (1995) Recruitment of Antarctic krill (Euphausia superba) and possible
causes for its variability. Mar. Ecol. Prog. Ser. 123, 45±56.
Siegel, V., Loeb, V. & GroÈger, J. (1998) Krill (Euphausia superba) density, proportional and
absolute recruitment and biomass in the Elephant Island region (Antarctic Peninsula)
during the period from 1977 to 1997. Polar Biol. 19, 393±98.
Stammerjohn, S.E. & Smith, R.C. (1997) Spatial and temporal variability in west Antarctic
sea-ice coverage. In Biology of the Antarctic Peninsula Area (R.M. Ross, E.E. Hofmann &
L.B. Quetin, eds). Antarctic Research Series No. 70, pp. 81±104. American Geophysical
Union, Washington.
Statistical Bulletin (1990a) Volume 1 (1970±1979). CCAMLR, Hobart. 61 pp.
Statistical Bulletin (1990b) Volume 2 (1980±1989). CCAMLR, Hobart. 109 pp.
Statistical Bulletin (1991) Volume 3 (1981±1990). CCAMLR, Hobart. 119 pp.
Krill Fisheries in the Southern Ocean
337
Statistical Bulletin (1992) Volume 4 (1982±1991). CCAMLR, Hobart. 133 pp.
Statistical Bulletin (1993) Volume 5 (1983±1992). CCAMLR, Hobart. 123 pp.
Statistical Bulletin (1994) Volume 6 (1984±1993). CCAMLR, Hobart. 143 pp.
Statistical Bulletin (1995) Volume 7 (1985±1994). CCAMLR, Hobart. 147 pp.
Statistical Bulletin (1996) Volume 8 (1986±1995). CCAMLR, Hobart. 150 pp.
Statistical Bulletin (1997) Volume 9 (1987±1996). CCAMLR, Hobart. 157 pp.
Statistical Bulletin (1998) Volume 10 (1988±1997). CCAMLR, Hobart. 134 pp.
Stasenko, V.D. (1967) Determining the rational krill fishing methods and the commercial
effectiveness of the chosen fishing gear. In Soviet Fishery Research on Antarctic Krill (R.N.
Burukovskiy, ed.), Document (TT 67-32683), pp. 61±78. US Dept. of Commerce, Washington, D.C.
Sushin, V.A. & Myskov, A.S. (1993) Location and intensity of the Soviet krill fishery in the
Elephant Island area (South Shetland Islands), 1988/89. In Selected Papers Presented to the
Scientific Committee of CCAMLR, 1992, pp. 305±335. CCAMLR, Hobart.
Trathan, P.N., Agnew, D.J., Miller, D.G.M., Watkins, J.L., Everson, I., Thorley, M.R.,
Murphy, E.J., Murray, A.W.A. & Goss, C. (1993) Krill biomass in Area 48 and Area 58:
Recalculations of FIBEX data. In Selected Papers Presented to the Scientific Committee of
CCAMLR, 1992, pp. 157±81. CCAMLR, Hobart.
Trathan, P.N., Everson, I., Miller, D.G.M., Watkins, J.L. & Murphy, E.J. (1995) Krill biomass
in the Antarctic. Nature 373, 201±202.
Trathan, P.N., Everson, I., Murphy, E.J. & Parkes, G.B. (1998) Analysis of haul data from the
South Georgia krill fishery. CCAMLR Sci. 5, 9±30.
Trivelpiece, W., Trivelpiece, S.G. & Volkman, N.J. (1987) Ecological segregation of AdeÂlie,
gentoo and chinstrap penguins at King George Island, Antarctica. Ecology 68, 351±61.
Vagin, A.V., Makarov, R.R. & Menshenina, L.L. (1993) Diurnal variations in biological
characteristics of krill, Euphausia superba Dana, to the west of the South Orkney Islands,
24 March to 18 June 1990 ± Based on data reported by a biologist observer. In Selected
Papers Presented to the Scientific Committee of CCAMLR, 1992, pp. 201±222. CCAMLR,
Hobart.
Watters, G. & Hewitt R.P. (1993) Alternative methods for determining subarea and local
area catch limits for krill in Statistical Area 48. In Selected Papers Presented to the Scientific
Committee of CCAMLR, 1992, pp. 237±49. CCAMLR, Hobart.
WoÈehler, E.J. (compiler) (1993) The Distribution and Abundance of Antarctic and Subantarctic Penguins. Scientific Committee for Antarctic Research, Cambridge. 76 pp.
Chapter 13
Management of Krill Fisheries in
Canadian Waters
Inigo Everson
13.1
British Columbia krill fishery
An experimental krill fishery began around 1970 off the east coast of Vancouver
Island, British Columbia the annual catch from which, up to 1987, remained below
200 t. Since that time the fishery increased to a peak in 1990 of over 500 t with a
value of Can$ 415 000. Annual catches are shown in Fig. 13.1. In 1992 some poor
quality krill were sold with the result that the market for British Columbian krill
became depressed and demand went down. A direct result was that landings in 1993
were very low (Nicol & Endo, 1997). Since that time landings have increased and in
Fig. 13.1
338
Reported catches of krill for the krill fishery in British Columbia.
Krill Fisheries in Canadian Waters
339
1994 were over 300 t with a value of Can$ 259 000 (Nicol & Endo, 1997, Pitcher &
Chuenpagdee, 1996, Ichii, Chapter 9).
Fishing has been carried out by two types of vessel, small trawlers that have only a
limited freezing capacity of around 5 t of krill a day and seine netters which land
large quantities of krill for processing and freezing ashore (Nicol & Endo, 1997,
Ichii, Chapter 9). This form of fishing practice effectively restricts the fishery to the
coastal region where small vessels can operate and larger vessels can return to port
within a few hours of making catches.
Currently the fishery is closely controlled and restricted to the Jervis Inlet, a small
and narrow inlet which opens into the Malaspina Channel situated on the eastern
side of the Strait of Georgia and shown in Fig. 13.2. The area is further subdivided
into a series of five management units which are designated by narrowing of the
channel or other topographical features.
In 1976 a total allowable catch (TAC) was set at 500 t for a fishing season
extending from November through to March. The criterion for setting the TAC was
that it should be no more than 3% of the estimated consumption of krill by all
predators in the Strait of Georgia (Nicol & Endo, 1997). The status of the fishery
and the justification for reviewing the TAC were discussed in 1995 at a workshop
meeting in Vancouver (Pitcher & Chuenpagdee, 1996) and the main questions
raised at that meeting concerned the ecosystem approach to management, how that
might be applied to the region and issues associated with funding the research
needed to develop this management advice.
13.2
Nova Scotian krill fisheries
Research studies investigating the feasibility of harvesting krill from the Nova
Scotian region began in 1972 arising from which a proposal was made to harvest krill
(Sameoto, 1975). Initially estimates of biomass were made using nets although
recent studies using acoustics have indicated a standing stock of krill in the Gulf of
St Lawrence of 8 to 96 kt (Simard & Lavoie, 1999, Everson, Chapter 3).
The Maritimes Region of the Department of Fisheries and Oceans Canada
(DFO) in 1991 issued its first scientific permit to harvest zooplankton. That licence
was for both krill and copepods, in the Gulf of St Lawrence. Exploratory fishing
commenced in November 1993 in the St Lawrence Estuary off Ste-Anne-des-Monts.
In 1994 so as to exercise control over any possible expansion in the fishery a TAC
was set at 100 t for krill and 50 t for copepods, specifically Calanus. During that
season the reported catches were only of 6.3 t of krill and 0.4 t of Calanus. In 1995
the TACs were increased to 300 t of krill and 2000 t of Calanus but catches were
again very low, 2 t of krill and 1 t of Calanus (Harding, 1996).
In late 1995, DFO received a proposal to develop a 1000 t experimental fishery for
krill, a forage species, on the Scotian Shelf and in the Bay of Fundy (Anon., 1996).
This proposal was reviewed through the regional advisory process (RAP) whose
report indicated that such a fishery was likely to have only a negligible effect on the
50 N
52 N
54 N
132 W
126 W
Longitude (degrees)
129 W
123 W
Malaspina
Strait
120 W
40'
44'
Outside
48'
52'
56'
50 N
12'
Longitude (degrees)
124 W
Jervis Inlet region
6'
Hotham
54'
48'
Centre
Inside
Middle
Fig. 13.2 The Jervis Inlet region showing the relationship of the fishery area to the coast of British Columbia. The five management
areas within the inlet are marked.
48 N
Latitude (degrees)
56 N
Latitude (degrees)
340
Krill: Biology, Ecology and Fisheries
Krill Fisheries in Canadian Waters
341
ecosystem. At that time there was concern that although the 1000 t TAC might have
no adverse effect on the ecosystem, should the fishery prove successful, there might
be pressure to expand. The consequences of such expansion needed careful consideration before approval for such action might be given. At this point the debate
was broadened to include what are now termed `forage species'; these are species
that form the diet of higher trophic levels. Arising from these considerations the
RAP Steering Committee organised a workshop to consider the broader implications of fisheries on forage species. The broad objective of the workshop was to
provide guidance for the future development of forage species fisheries in the DFO
Maritimes Region. That workshop took place in April 1997 in Halifax (Head, 1997)
and was attended by both national and international experts as well as local interested parties.
13.3
Krill fisheries management
The experiences from both the Atlantic and Pacific Canadian krill fisheries have
been running in the same direction and it is convenient to consider the two together.
During the discussion phase of the 1995 Vancouver Workshop (Pitcher &
Chuenpagdee 1996) a series of questions was identified by J. Morrison of DFO that
needed to be addressed in order to develop revisions to the management plan. The
first was aimed at the industry and asked specifically whether it wished to support
future surveys in order to provide information that might be used to revise the TAC.
The second question was aimed at scientists and managers seeking to determine the
minimum information that would be required to revise the current TAC. The third
question was directed at ecological modellers encouraging them to come up with
models that take account of ecosystem interactions and also the risk associated with
different management strategies.
In the course of the discussion two points emerged, firstly that an ecosystem
approach, whereby dependent species were factored into the equation, was essential. The second point related to the resources available to provide assessment
advice, a question that, put simply, reduces to comparing the value of the krill
fishery to that of higher priced finfish such as salmon and hake. These points and
other matters arising from the discussion were taken forward when the proposal to
increase the krill TAC on the Atlantic seaboard was considered.
The 1997 Halifax Workshop (Head, 1997) considered key questions related to
the overall objectives for forage fisheries in Canadian waters the first of which was:
What should be the conservation objectives related to fisheries on forage
species?
This question can be looked at in the following ways:
. maintenance of ecological relationships/ecosystem integrity;
. minimisation of risk of irreversible decline;
342
Krill: Biology, Ecology and Fisheries
. maintenance of ecosystems and constituent species within the bounds of natural
fluctuation;
. maintenance of full recruitment potential (including genetic diversity); and
. allowance of fisheries which meet conservation objectives (and maximize
knowledge returns).
Many of these topics are essentially a restatement of the requirements of Article
II of the CCAMLR Convention that is operational in the Southern Ocean and is
discussed in detail by Miller and Agnew (Chapter 12). The key difference is associated with the emphasis of the different components. In the Southern Ocean few of
the dependent species are the subjects of commercial harvesting so that the fishery
there is being targetted almost exclusively at one trophic level, that of the krill. By
contrast the Canadian situation is one where many of the dependent species are
themselves the subject of significant commercial fishing. Thus the following fish
species, discussed by Tanasichuk (Chapter 7), are dependent to a greater or lesser
extent on krill in Canadian waters: salmon (Onchorynchus tsawyscha and O.
kisutch), hake (Merluccius bilinearis, M. productus and Urophycis chuss), cod
(Gadus morhua), haddock (Melanogrammus aeglefinus), herring (Clupea harengus
and C. pallasi), capelin (Mallotus villosus) and redfish (Sebastes spp). The management regime in this forum needs to take account of the status of the dependent
species and furthermore come to a view as to the likely effect of different levels of
krill harvest. Such answers have an economic as well as ecological connotation.
The second set of questions raised at the Halifax Workshop related to the
requirements that would need to be met by management of the fishery in order for
the conservation objectives to be satisfied. The conclusions that arose from this were
summarised in the following key points:
. The knowledge base must be adequate to allow risk evaluation.
. There should be confidence that monitoring and controls are adequate to allow
management decisions (i.e. responses to change) to be made and implemented in
timely fashion.
. Monitoring of direct impact and indirect impact on dependent species and of bycatch is required.
. Ongoing assessment of natural levels of fluctuations in species and ecosystem is
required.
. Proponents should bear the incremental costs of acquiring knowledge and
managing the fishery.
Arising from this it was clear that the CCAMLR approach (Miller & Agnew,
Chapter 12) provided a good framework against which to consider krill fisheries on
the Scotian shelf. However in taking the matter forward there was a clear indication
that that approach would need adjustment in order to address the specific
requirements of Canadian fisheries.
These adjustments were not required for the krill-centred part of the system
Krill Fisheries in Canadian Waters
343
where the CCAMLR approach was seen as providing a good model to use but for
the dependent species that were subject to a commercial harvest. The developments
that were required were envisioned in the report of the RAP Steering Committee.
That Committee recommended, amongst other things, that:
. Decisions made concerning these (forage) fisheries and their scientific justification must be credible to a broad range of stakeholders.
and
. The CCAMLR approach requires ongoing monitoring of the biological health of
dependent species. In the case of krill, one such species would be hake. Steps
would need to be taken to identify dependent species and initiate monitoring
programmes for any emerging or existing forage fisheries.
The term `stakeholder' is an important term in this context because it draws
together those with interests in harvesting krill, harvesting fish which feed on krill,
with those interested in maintaining a healthy ecosystem in perpetuity. The aims
and aspirations of these separate and sometimes opposing interests need to be
considered by the fisheries' managers as they develop policy. Where these recommendations differ from the CCAMLR approach is governed by two factors. Firstly
the Canadian fisheries fall under national jurisdiction so that policy can be made,
implemented and monitored; essentially this is funded through the licence revenue
from the fisheries. The Canadian Government can make the rules and implement
them. CCAMLR by contrast operates by consensus and is an international management organ; its income derives from payments by member states and these
member states are responsible for ensuring that their own nationals comply with
conservation measures. The second point is that many of the dependent species in
Canadian waters are themselves subject to commercial harvest whereas those in the
Southern Ocean, with the exception of the mackerel icefish, are not fished. It is this
second point which is worth considering in a little more detail.
The two criteria that CCAMLR have agreed for the Antarctic krill fishery are
that over a 20 year period the minimum spawning stock biomass should not fall
below 20% of its unexploited level in any season, and at the end of the period, that
the mean spawning stock biomass should not fall below 75% of its pre-exploitation
level. These criteria were established in order to provide a precautionary catch limit
that satisfied the requirements of Article II of the CCAMLR Convention. In the
Canadian context, satisfying the requirements of Article II is only part of the story
because reducing the mean spawning stock biomass will most likely have an effect
on commercial fish species.
The scientific advice that would be sought in order to address these points focuses
on the dependent species because information is then required on the effect of a
25% reduction in krill availability to, for example, hake. A likely outcome of such an
exercise would be that hake production, and consequently its TAC, would go down.
From the point of view of the stakeholders who are interested in fishing this reduces
344
Krill: Biology, Ecology and Fisheries
to the simple economic question of whether it is more profitable to fish for krill or
hake.
Energetically, based on a simple `rule of thumb', we might assume that ten tonnes
of krill lead to the production of one tonne of hake. If one assumes a conversion
ratio of perhaps 10% for the production of krill meal or 50% for krill products for
direct human consumption on the one hand and perhaps 60% for headed and gutted
hake then we have a basis for comparing tonnages. These can then be considered
alongside the market value of the different products in assessing the value of the
different fisheries. Further development of these ideas is clearly necessary but falls
outside the bounds of the current chapter.
In summary the Canadian approach has been one to impose a high level of
caution from the outset specifically to address ecosystem considerations. In developing the approach the experiences of CCAMLR have been accepted and are being
further developed to consider the implications of harvesting at different trophic
levels in order to develop a sustainable management regime that provides a satisfactory return to stakeholders.
References
Anon. (1996) Krill on the Scotian Shelf. DFO Atlantic Fisheries Stock Status Report 96/106E.
Harding, G.C.H. (1996) Ecological factors to be considered in establishing a new krill fishery in
the Maritimes Region. DFO Atlantic Fisheries Research Document 96/99, 11 pp.
Head, E. (1997) Proceedings of the workshop on Ecosystem considerations for krill and other
forage fisheries. Canadian Stock Assessment Proceedings Series 97/5, 61 pp.
Nicol, S. & Endo, Y. (1997) Krill fisheries of the world. FAO Fish. Tech. Paper 367, 1±100.
Pitcher, A.J. & Chuenpagdee, R. (eds) (1996) Harvesting krill: ecological impact, assessment,
products and markets. Fisheries Centre Research Reports 3 (3). Fisheries Centre, University of British Columbia, Canada. 82 pp.
Sameoto, D.D. (1975) A krill fishery in the Gulf of St Lawrence. Can. Fisherman Ocean Sci.
61, 16±17.
Simard, Y. & Lavoie, D. (1999) The rich krill aggregation of the Saguenay±St Lawrence
Marine Park: hydroacoustic and geostatistical biomass estimates, structure, variability and
significance for whales. Can. J. Fish. Aquat. Sci. 56, 1182±97.
Chapter 14
Krill Fisheries and the Future
Inigo Everson, David Agnew and Denzil Miller
The foregoing chapters have brought together much information on the biology,
ecology and fisheries for krill in different parts of the world. Against that background we can now reflect where that information is likely to lead in the future. We
consider this from the several different perspectives.
We have seen in Chapter 10 that there are several main types of product derived
from krill which can be categorised by their market value and volume. On the one
hand we have the lower value high volume products such as fish-meal and aquaculture feed and for which the technology is quite well established. At the other
extreme are products for direct human consumption and for these there is a
requirement for krill of a specified quality to be processed in an efficient manner.
This leads to low volume high quality products. In addition there are the products
that are produced as by-products from one of the other processing systems. Byproducts themselves are unlikely to have much influence in guiding the fisheries
except by offering a marginal increase in revenue so we need not consider these
further in a consideration of future developments. High value low volume products
tend to be produced as a result of some key technological development; this means
that any developments in this line are more likely to appear in Patent files than texts
such as this one. The high volume products and in particular those supporting the
aquaculture industry can provide further food for thought.
Over the past thirty or so years there has been a rapid increase in the aquaculture
industry in many parts of the world which has resulted in a requirement for large
amounts of feed. Krill products have been used increasingly for this purpose largely
because of their nutritional content and also because the carotene pigments enhance
the colour of the farmed fish. It is interesting to note that the main centres of krill
fishing provide a poor match for the regions where aquaculture is at its greatest. The
aquaculture industry is particularly strong in northern Europe, north America and
south-east Asia in the northern hemisphere and yet the only major krill fishery is in
Japan. In the southern hemisphere the largest krill fishery is in the Southern Ocean,
a region not amenable to the needs of aquaculture, whereas the main aquaculture
interests appear to be centred in Tasmania and Chile.
Krill concentrations are known to occur close to many of the marine locations
where the aquaculture industry has developed and it is worth considering why more
use is not made of the local krill. The answer lies probably in the seasonal cycle of
345
346
Krill: Biology, Ecology and Fisheries
the krill. Aquaculture requires a consistent supply of feed year-round and this is
unlikely to come from a local source that has large-scale seasonal cycles of abundance. Improvements in the efficiency of bulk transport mean that sufficient feed
can be transported over great distances and, with careful planning, this can smooth
out variation in krill supply to ensure a steady supply of feed. Thus the main
advantage to be gained from using local krill for aquaculture feed is only when there
is a need for the krill to be fresh. In spite of this there are concentrations of krill
close to areas where aquaculture is a key industry such as Scotland and Norway
where Meganyctiphanes norvegica is seasonally present and likewise for Nyctiphanes australis off Tasmania (Nicol & Endo, 1999). That fisheries have not
developed in these areas may in part be due to the ready supply of krill products
from the established fisheries but also may be a result of considerations for fisheries
management such as are currently under discussion for the Canadian fisheries.
The debate on the Canadian krill fisheries has taken the ecosystem approach of
CCAMLR a step further by extending the philosophy to encompass dependent
species that are themselves the subject of commercial fishing. Although in the
background, due to the concerns over the impact of Antarctic whaling, when the
CCAMLR Convention was developed the current scenario of dependent species
being harvested is of marginal concern in that forum. The only example currently
is the mackerel icefish Champsocephalus gunnari (Miller & Agnew, Chapter 12).
The Canadian approach has introduced the more complex consideration of harvesting at two trophic levels within the same system. In order to achieve this
there are the considerations of maintaining populations of the harvested and
dependent species sustainably and in perpetuity, the current CCAMLR approach,
allied to assessments of the relative values of the products from the different
trophic levels (Everson, Chapter 13). That part of the equation which relates to
the CCAMLR approach can be developed effectively by a scientific analysis of
the system; this has been largely achieved and the approach is being refined
(Miller & Agnew, Chapter 12).
Harvesting at more than one trophic level can also introduce socio-economic
factors into the equation. At the economic level, this might be viewed as a simple
trade-off between the values of individual products adjusted for volume and production costs. The social component comes in when a desire is indicated to harvest
krill and to do so would entail a reduction in total allowable catch (TAC) of a
`traditional' finfish species. The resolution of that situation requires scientific advice
on the impact of different levels of harvesting at the different trophic levels leading
to a political decision to decide how the division should be made. This is in effect a
restatement of the fact that it is not possible to harvest two species from different
trophic levels at their individual levels of maximum sustainable yield (MSY)
(Beddington & May, 1980). The political decision determining which fishery will be
permitted the greater proportion of its individual MSY will very likely be further
affected to take account of lobbying groups whose primary concerns might not be
associated directly with fishing. Such considerations might include, for example, the
Krill Fisheries and the Future
347
tourist potential from whale watching or the scientific issues associated with
maintaining biodiversity.
Two points arise from these considerations. Firstly, there is the need to understand the interactions between the species that are being harvested at the two
trophic levels. The second point indicates that there is likely to be more scope for
krill fishing when there is no harvesting at higher trophic levels.
Understanding the interactions between different trophic levels, particularly
when that information is likely to be taken forward in determining management
policy, is a daunting task. In the case of CCAMLR and its development of an
Ecosystem Monitoring Programme (CEMP) indicator species and sampling regimes
had to be developed de novo. This has meant that it has taken over a decade to
develop the database sufficiently to be able to begin to understand the intricacies of
the situation. Ironically when the dependent species is a commercial fish there is
likely to be a database already established. The fishery can also provide information
on total catch, catch per unit of effort (CPUE) and biological information on the
species in question. Fish stock assessments traditionally require estimates of
standing stock, recruitment, growth and mortality. If the species is of significant
commercial importance then other ecological components such as diet, spawning
cycles, condition, distribution and migrations will also have been studied albeit with
less intensity. Much information may already be available therefore with which to
study trophic interactions. Examples taken from the chapters of this volume where
this approach might be considered are hake in Canadian waters and walleye pollock
and blue eye in Japanese waters (see Tanasichuk, Chapter 7 and Endo, Chapter 7).
Returning to the potential for expansion in krill fisheries, it is clear that, where the
intention is to harvest at two trophic levels, the forage fishery scenario, the process
of developing a management framework is more complicated than that for the
CCAMLR type system. The scope for expansion envisioned by the difference
between dependent species' consumption of krill and the current fishery in the
Southern Ocean means that, should there be a desire to expand the world krill
fishery, there is capacity within the CCAMLR region.
Integrating krill fisheries into the general category of forage fisheries, as has been
done in Canada and Alaska (there is currently a prohibition on exploiting forage
fish in Alaskan waters), provides for some interesting avenues for future consideration of fisheries on mid-trophic level animals. In some ways krill are similar to
squid and small pelagics, both of which are subject to successful fisheries in many
parts of the world. These animals are characterised by having a short life-span and a
seasonal migration pattern that is largely driven by oceanographic circulation (e.g.
Agnew et al., in prep). Consideration of these animals as food for dependent toplevel predators (mammals and birds) has thus been extended to other exploited
species (i.e. fish) which also depend upon them. There is still a considerable way to
go in generating suitable management regimes for these mid-trophic level species,
since the systems under consideration usually respond to change in very complex
ways.
348
Krill: Biology, Ecology and Fisheries
Even the simplest apparent causal relationship, such as that of the effects of the
Alaskan pollock fishery on Steller sealion populations (Lowell et al., 1995), or the
North Sea sand eel fishery on seabirds (Wright, 1996) is often extremely difficult to
confirm. For instance, in the case of sand eels, it appears that recruitment variability,
which appears to be independent of the adult fishery, has caused the observed
fluctuations in seabird breeding success. Nevertheless, the underlying issues are
often similar for different fisheries, and the work done in establishing management
regimes for krill can arguably be seen as providing useful information for consideration in other fora. Likewise there are lessons to be learnt for krill fishery
managers by careful scrutiny of the practices in other fisheries. This is true particularly where harvesting is occurring simultaneously at two or more trophic levels
and the problems are compounded by the addition of environmental and social
concerns regarding the effects of such harvesting on all trophic levels within an
ecosystem. A need to move towards `holistic' fisheries management, considering the
effects of harvesting on the ecosystem as a whole, is likely to assume much more
prominence in the 21st century, and is likely to have a significant effect on the way
krill and other fisheries are managed.
References
Agnew, D.J., Hill, S. & Beddington, J.R. (in prep) Predicting the recruitment strength of an
annual squid stock: Loligo gahi around the Falkland Islands.
Beddington, J.R. & May, R. (1980) Maximum sustainable yields in systems subject to harvesting at more than one trophic level. Math. Biosci. 51, 261±81.
Lowell, F.W., Ferrero, R.C. & Berg, R.J. (1995) The threatened status of Steller sea lions,
Eumetopias jubatus, under the Endangered Species Act: Effects on Alaska groundfish
fisheries management. Mar. Fish. Rev. 57, 14±27.
Nicol, S. & Endo, Y. (1999) Krill fisheries: development, management and ecosystem
implications. Aquat. Liv. Res. 12, 105±120.
Wright, P.J. (1996) Is there a conflict between sandeel fisheries and seabirds? A case study at
Shetland. In Aquatic Predators and their Prey (S.P.R. Greenstreet & M.L. Tasker, eds),
pp. 154±65. Fishing News Books, Oxford.
Glossary
abdomen
abiotic
aggregation
Alcidae
allometry
Amphipoda
Antarctic Polar Front
antioxidant
austral
autodigestion
autoproteolysis
baleen whale
beam trawl
benthic
benthopelagic
biomass
biotic
boreal
posterior part of the body
pertaining to non-living things
a large concentration of krill which may represent a group
of swarms
family of birds belonging to the order Charadriiformes
which includes auks, puffins, razorbill and guillemots
the relative growth rate of part of an organism in relation to
another part or to the whole
laterally compressed Peracarid Crustacea in which the
cephalothorax lacks a carapace and the whole head is fused
to one, rarely two, thoracomeres
the frontal zone at the northern extremity of the Southern
Ocean. Recognised by an increase in surface water temperature of a few degrees as it is crossed in a northerly
direction
a compound added to food products to reduce the effects of
oxidation and the accompanying degradation of the product
at high latitude in the Southern Hemisphere (cf. boreal)
post mortem degradation due to the action of enzymes
present within the body
use of natural proteolytic enzymes to produce a high yield
protein concentrate
the group of whales characterised by having horny plates
derived from the mucous membrane of the palate which act
as a strainer to filter plankton from the water
a frame net having a beam to hold the net open during
fishing operations
living on the seabed
mainly associated with the seabed but making excursions
up into the water column
the total mass of biological material of either a species or a
group present in an area. Synonymous with standing stock
relating to life or living creatures
of the north. The boreal zone is the region of short summers and long cold winters
349
350
Krill: Biology, Ecology and Fisheries
brood pouch
carapace
cephalothorax
chitin
chumbait
C : N ratio
cod end
cohort
condition index
Copepoda
combined standardised index
critical period distance
crystalline cone
cytoplasm
Decapoda
deep scattering layer
demersal
diel
Discovery Investigations
diurnal
dual beam transducer
echo-chart
echogram
exoskeletal shield to protect eggs or larvae
an exoskeleton shield covering part of the dorsal surface
the region of the body formed by the fusion of the head and
thorax
a nitrogenous polysaccharide with the formula (C8H13N5)n
occurring as skeletal material in crustacea
ground-bait thrown into the water by fishers working with
hook and line fishing devices
carbon to nitrogen ratio. Used as an indicator of the
amount of protein present in a tissue
the narrow end of a trawl net in which the catch is retained
group of animals arising from a single spawning event. If
the spawning events are annual these correspond to year
classes
the ratio of measured mass to estimated mass; the estimated mass coming from an overall long-term dataset
a Subclass of the order Crustacea
an indication of the availability of krill to dependent species and obtained by combining several CEMP indices
from one season (CCAMLR)
a CCAMLR index to indicate the degree of overlap
between fishing activity and dependent species foraging
activity
the outer refractive body of a crustacean eye which acts as a
light guide
that part of the cell which exists outside of the nucleus but
inside the cell wall
an order of Malacostracan Crustacea with 3 pairs of thoracic limbs modified as maxillipeds and 5 as walking legs
(shrimps, prawns, crabs, lobsters etc.)
an oceanic layer characterised by a high volume backscatter and showing up as an echo-trace
living on or near the seabed
synonymous with diurnal in the sense that it pertains to a
period of 24 hours, not in the sense of by day rather than by
night
research programme undertaken by the UK during the
inter-war period mainly in the Southern Ocean to study
whales and their environment
activity over a 24-hour time-scale. Comparison of day and
night activity
echo-sounder transducer having two co-axial beam patterns, a narrow and a wide beam. Targets seen in the narrow beam are assumed to be on the axis of the wide beam
paper output from an echo-sounder
as for echo-chart
Glossary
echo-sounder
echo-integration
echo-trace
ejaculatory ducts
El NinÄo
endogenous
epipelagic
exogenous
exopodites
exoskeleton
eye stalk
facultative schooler
fecundity
forage fishery
gadoid
geometric scattering
geostatistics
germinal zone
gravid
growth
gyre
hepatopancreas
351
device for sending pulses of high frequency sound into the
water and listening for return echoes. The time delay to
receipt of the echo provides an indication of target range
accumulating echo signals over a period of time. The
integrated signal is assumed proportional to the density of
the targets
marks on an echogram to indicate the distribution by depth
and time of targets
a narrow muscular tube forming the lower part of the vas
deferens and leading into the copulatory organ (petasma)
a complex set of changes in water temperature in the
Eastern Pacific Equatorial region producing a warm current; it occurs annually to some degree between October
and February
a general term used to suggest that the causes of some
physical condition are due to internal factors
in the upper part of the pelagic zone, i.e. near to the sea
surface
resulting from causes external to an organism
the outer ramus of a crustacean appendage
hard supporting or protective structures that are external
to and secreted by the ectoderm
a paired stalk arising close to the median line on the dorsal
surface of the head of many Crustacea, bearing an eye
having the ability to live and adapt to schooling while not
being restricted to that mode of life
the number of young or eggs produced by a species or
individual
fishery based on species that form the food of other species
higher up the food chain
member of the cod family
the situation where an acoustic target is large relative to the
wavelength of the sound such that there is little change in
target strength with frequency
branch of statistics dealing with the analysis of data of
highly contagious distributions
a layer of epithelium from which ooÈcytes develop
carrying ripe eggs, ready to spawn
an irreversible change in an organism accompanied by the
utilisation of material and resulting in a change of volume
circular pattern of water movement generally arising due
to topography or else the counter movement of two water
masses
a glandular diverticulum of the mesenteron, believed to
carry out the functions of the liver and pancreas of higher
vertebrates
352
Krill: Biology, Ecology and Fisheries
hydrolysate
ice-algae
instantaneous growth rate
isolume
iteroparous
krill flux
length-at-age
length frequency distribution
length density distribution
lipid
lipofuscin
longevity
luminous organs
mode
mortality
moult
Myctophidae
Mysidacea
nearest neighbour distance
neritic
net selectivity
net-sounder
ooÈcyte
ooÈgonia
ova
ovary
pack-ice
pelagic
perivitelline space
a krill product produced by controlled autoproteolysis and
thermal precipitation of the water soluble protein and thus
extract it from the exoskeleton
algae which grow on and in pack-ice
growth rate determined over a short period of time such as
a day or week
the same light level
reproducing more than once during a lifetime
rate of movement of krill through a region
a table to show the range of length of krill of a particular age
frequency distribution of lengths of krill obtained from a
sample or group of samples
length frequency distribution of a population and calculated as a weighted length frequency distribution from all
relevant samples
any of a group of fat or fat-like substances
a fatty hydrocarbon substance that accumulates in certain
animal tissues upon ageing
life span
light producing organs
the value represented by the greatest number of individuals
in a frequency distribution
generally applied to the rate at which individuals in a
population die. Frequently described by the formula: Nt =
N0 e±Zt where N is the number at times t and zero and Z is
the coefficient of total mortality
the act of casting off the outer layer of the integument
Lantern fishes
Opossum shrimps, an order Pericarida, carnivorous
shrimp-like crustaceans having biramous thoracic appendages
the distance between adjacent krill in a swarm
the shallow-water coastal region
the degree to which animals of different sizes are retained
by a fishing net
an echo-sounder operating from a fishing net to determine
the depth of the net from the surface or seabed and also the
amount of fish entering it
female gametocyte
an egg-mother-cell
eggs or egg-cells
female gonad
ice formed by freezing of the sea-surface
in the water column, i.e. above the seabed
the space between the vitelline membrane and the capsule
of an egg
Glossary
petasma
photoperiod
photophore
phototaxis
phytoplankton
plankton
pleopod
polymodal
polyna
Procellariiformes
proteolysis
pycnocline
Rayleigh scattering
recruitment
rostrum
salp
schooling
sea-ice
selectivity
setae
shelf
shelf-break
spermatophore
split-beam transducer
Southern Ocean
standing stock
353
male external sexual character, situated on the first pair of
abdominal pleopods
light to dark regime in which an animal lives
light producing organ
movement in response to light
microscopic drifting plant life
drifting life. Essentially living organisms that are unable to
make significant movements against oceanic currents
an abdominal appendage adapted for swimming
a frequency distribution having more than one peak
area of open water within the pack-ice zone
an order of birds including albatrosses, petrels, shearwaters and fulmar, characterised by tube-like nostrils on
their beaks
the degradation of proteins into peptides and amino acids
by cleavage of their peptide bonds
density discontinuity
the situation where the strength of the echo from an
acoustic target is strongly proportional to the size of the
target or else is in a resonant region where echo strength
varies strongly with target size
first arrival of juvenile or young individuals into the
population. In fishery terms arrival into the exploited
phase of the population
a median anterior projection of the carapace
a transparent pelagic member of the order Tunicata (sea
squirts)
the process by which krill aggregations and swarms are
formed
generally synonymous with pack-ice
characteristics of a net describing the proportion of individuals of different sizes that are retained by a net
long thin bristle-like structures on the end of euphausiid
limbs
relatively shallow area adjacent to a land mass
the transition zone where the shelf ends and the water
becomes very much deeper in the open ocean
a packet of spermatozoa enclosed within a capsule
an echo-sounder transducer which is divided into four
quadrants. Differences in phase between return echoes
from the same target are used to determine the position of
the target on the main axis of the transducer
the area of ocean surrounding the Antarctic Continent.
The northern limit is generally taken to be the Antarctic
Polar Front
synonymous with biomass
354
Krill: Biology, Ecology and Fisheries
stock
submersible
super-swarm
swarm
sympatric
target strength
telson
testis
thelycum
thorax
topographic
total allowable catch
transect
triacylglycerol
unimodal
upwelling
uropod
vasa deferentia
vitellogenesis
volume backscattering strength
warm-core rings
zooplankton
group of animals of a single species which can be treated as
a single unit for management purposes
instrument package for in situ sampling of the underwater
environment
extremely large swarm of krill
dense concentration of krill which appear to be organised
and/or moving in a coherent manner
two species or populations having common or overlapping
geographical distributions
an acoustic ratio describing the proportion of incident
sound that is reflected back by a target
the post-segmental region of the abdomen
male gonad
female external sexual character
the region of the body lying between the head and abdomen. Often used erroneously as a synonym for cephalothorax
detailed description of the seabed
fishery term to limit the total catch of a species within a
management area
line sampled during a survey
a triglyceride. The product of the esterification of a glycerol molecule with three fatty acid molecules
a frequency distribution having a single peak
water circulation pattern which brings deep water to the
surface
an appendage of the abdominal somite preceding the telson
ducts leading from the testes to the petasma
the process of yolk deposition in the ova
acoustic term to describe the proportion of the incident
energy that is reflected by a layer or volume of seawater
areas of warm water that are entrapped within cooler water
masses by local circulation patterns
animal plankton
Index
abdomen 2
abundance 11, 15, 26, 29, 210, 252, 256, 315, 321,
328
larval 53
numerical 59, 67, 70
acoustic 20±32, 48, 54±5, 72±3, 80±98, 294
beam pattern 21±24, 27
acoustic calibration 23, 30, 89
acoustic density 89
acoustic detection range 239
Acoustic Doppler Current Profiler (ADCP) 73,
94
acoustic
Echotrace 249
frequencies 22, 23, 25, 27
sampling 8, 10, 11, 13, 20
scattering types 25
layer 13, 54, 59, 93, 94
signal intensity 21, 27
sonar 13, 20, 22, 96, 238, 239, 244, 247
survey 210, 223, 311, 316
systems, mltifrenquency (acoustics) 10, 25, 28,
48
tools 20±23
transducer 20±25, 30
Target Strength 20±2, 24, 26, 30, 48, 89, 90, 312,
314
difference 48
volume scattering 20, 22, 24, 25, 26
Admiralty Bay 117, 127
adult krill 37, 38, 42, 44, 69, 71, 105, 117
Advanced Very High Resolution Radar 247
aerial photography 56
Africa 234
age
based model 219, 312
class 127
density 74
determination 106±116
group 108, 109
maximum 113, 114, 133
pigment, fluorescent 115
age at
first capture 312
first reproduction 152
first spawning 150±152
maturity 112, 114, 151
recruitment 312
aggregation 11, 12, 44, 48, 54, 58, 60, 80±98, 160,
204, 205±6, 238, 239, 251, 253, 257, 322
benthopelagic 44, 45, 47
density 88±92
seasonal variation 94, 97
sex ratio 159
size 94
surface 42, 45, 158, 159
three-dimensional picture 96
Akita Prefecture 285
Alaska 157, 191
Alaska, Gulf of 3
Albatross 206
Black-browed 195±6, 204, 206, 315±6
grey-headed 195±6
light-mantled sooty 195±6
Alcidae 192, 204
alcohol content 263
alcohol fixation 128
Aleutians 3, 110, 111, 113, 114, 291
allometric relationship 132, 170
Ami-ebi 296
Amino acid 269, 270, 273, 295
ammonium excretion 126
Amphipoda 190, 209
Anchovy Peruvian 326
Animal feed, krill in 265
Antarctic 5, 14, 105, 121, 160, 228, 256
Antarctic Circumpolar Current (ACC) ix 68, 69,
155, 310
Antarctic cod, see marbled rockcod
Antarctic region 278
Antarctic krill, (see also Euphausia superba) 47,
106, 125, 137, 229±244, 262, 264±5, 267,
275±7, 294±5, Plate 1
Antarctic Krill Concentrate 266, 267
Antarctic Peninsula 69, 70, 91, 94, 95, 97, 109,
111, 112, 118, 119, 121, 124, 134, 137, 139,
155, 165, 166, 167, 217, 306, 312, 313, 318,
323
Antarctic Polar Frontal Zone (APFZ) 5, 64, 74,
198, 304
antenna 2, 33, 34
antennule 2, 33
Anticosti Island 58
Antioxidant 268
Anvers Island 94
Aomori Prefecture 45, 46
355
356
Index
aquaculture 5, 228, 252, 254, 264, 265, 267±274,
276, 278
feed(ing) 262, 295, 297
additives 269±270
wastage 269
stimulant 269
trials 265
fish feed colour 270
Aquafeed 268, 269, 272, 274, 278
aquaria 106, 158
Aquarium feed 252, 265, 274
Arctic 3, 110, 154, 157
Ocean 3
Argentina 304
arginine 271
assessment, single species 202, 220
assimilation rate 184
Astaxanthin 268, 269, 272
Atlantic Ocean
Antarctic sector 63±9, 109, 121, 140, 229, 234,
237, 238, 239, 241, 244, 306, 315
north 1, 2, 4, 10, 56, 112±17, 122, 123, 134, 157,
160, 204, 341
Auke Bay 157
auklet 187, 188
crested (Aethia cristaella) 186
rhinoceros (Cerorhinca monocerata) 182, 185,
186
Australian Antarctic Division 66
Australian National Antarctic Research
Expedition,
Australia, south east 267
Autolysis 39, 263, 266
Autonomous Underwater Vehicle, (AUV) ix 98
availability (of krill) 9, 12, 104, 206, 211, 220, 300
Available stock (of krill) 328
avoidance 9, 10, 11, 12, 13, 15, 26, 89, 104, 239
bacteria 263
bacterial flora 263
Baffin Island 4
bag-sampler 10, 88
bait. 262, 265, 285, 295
Baja California 107
balaenopterids 204
Baltic 4, 158
Barents Sea 4, 110, 112, 114, 124, 133, 134, 137,
153, 160
bathymetry 60
Bass, largemouth (Micropterus salmoides) 268,
270
Sea (Dicentrarchus labrax) 268
Beaufort Sea 3, 112, 191
Beaufort scale 239
Bechervaise Island 322
behaviour 28, 234
diel, 28
swimming/tilt angle 27, 48
behavioural
change 104
strategy 98
Belle-Isle Strait 57, 58
Bellingshausen Sea 90, 302
benthopelagic 41, 44, 45, 46
Bering Sea 3, 112, 118, 120, 185
bias 9, 10, 11, 15, 16, 26, 34, 108, 133
Biological
environment 138
observations 33±39, 258
biomass 11, 12, 48, 91, 185, 310, 311, 315, 317, 325,
339
density 96
pre-exploitation 203, 311, 343
trends 311
BIOMASS (Programme) ix 10, 47, 53, 55, 59, 60,
65, 69, 70, 89, 90, 195, 220, 304
Biscay, Bay of 4, 107
blue eye (Chlorophthalmus albatrossis (borealis)
182
body
growth 117
length 33, 242
size 232, 242
volume 172
boreal 110
bottom topography 75
Bouvet Island 302, 313
Bransfield Strait 68, 69, 72, 92, 94, 217
Bristol Bay 185
British Antarctic Survey, ix 13
British Columbia 1, 55, 124, 133, 134, 277, 338,
340
brittle stars 185
broods, multiple 168
Bulgaria 229, 230, 303
by-catch 342
Cabot Strait 57, 58
Calcium 273
California 107, 111, 112, 114, 118, 120, 122, 133,
134, 135, 136, 153
California current 3, 4, 54
calorie content 266
calyptopis 117
Canada 110, 304, 341±4, 346
East coast 56±61, 129, 157, 265, 339±341
West coast 53±6, 111, 120±1, 213, 228, 252, 254,
256, 267, 338±9
Capelin (Mallotus villosus) 133, 191, 342
Cape Inubo 46
carapace 2, 33, 34, 35, 38, 162
length 129±131, 132
carbohydrase 263
carnivore 122, 123, 128, 157
carotenoid pigment 266, 268, 269, 270, 272, 275
carp 271
catfish 271
catabolism, coefficient of (K) 118
catch
annual 244, 246, 251±8, 285, 288±9, 295±6,
301±2, 303, 305, 312, 327, 338±9
date, median 306, 307
Index
handling
fish pump 244, 251, 288
plastic container 251, 288, 294
history 312
levels 300, 314, 324
limit 294, 315
precautionary 203, 312±5, 318, 325, 329, 343
records 258
reported 230, 291
reporting of 313
statistics 300
total 228, 313, 322, 324, 325, 328
total allowable 55, 202, 310, 339, 343
trends 302, 306
trigger 312
value 339, 341
Catch per unit of effort, CPUE±239, 240, 241, 242,
249±52, 254±58, 286, 316±20, 325
Catch per unit offishing time (day or hour) 316,
319, 320, 324±5
search time 316, 317
catch rate,
daily 9, 234, 238, 247, 249, 253
prediction 320
catches 44±5, 64, 68, 236, 237, 252
CCAMLR 13, 16, 34, 37, 38, 64, 74, 75, 90, 194,
195, 202±07, 213, 215, 220, 229, 231,
300±330, 342, 343
Article II±309, 342
Ecosystem Monitoring Programme (CEMP)
199, 206±8, 211±2, 220±22, 320
central place forager 213
cephalothorax 2, 36, 263, 269, 276
chaetognath 43
chemical trail, mating 158
chemicals and pharmaceuticals 275±278
chemotherapy 267
Chile 229, 230, 303±4, 307, 319, 322, 329
China Sea 3, 297
Chitin 275
chlorophyll 58, 127, 157
chub mackerel, Japanese (Scomber japonicus)
182, 185, 187
chumbait 244, 295
circumpolar 4, 63
current 68, 69, 75
deep water (CDW) 160
cluster analysis 14
C : N ratio 126
coastal
current 72, 204
region 67, 204
cod 133, 191, 342
Arctic (Boreogadus saida) 190
Atlantic (Gadus morhua) 192
Pacific (Gadus macrocephalus) 47, 182, 183,
184, 187
codend 233, 251
cohort 55, 107, 108, 114, 118, 127, 133, 136, 138,
153
357
cold-core ring 41
cold-water species 107
colour of krill 38, 232, 245, Plate 2
green 232, 238, 240, 258, 263
pink 232
time of catch 268, 316
white (colourless) 240, 245
Combined Standardised Index 207, 208, 211,
221±2
commercial
fishing 1, 4, 34, 67, 68, 72, 74, 75, 76, 208, 211,
212, 223
fleets 68, 217
harvest 174, 343
net hauls 319
processing 38
trawl 10, 138, Plate 5
Commission for the Conservation of Antarctic
Marine Living Resources ± see CCAMLR
commissions, fishery 328
common squid, (Todarodes pacificus) 183, 185, 187
composite index (hauls) 14
composite index (krill abundance) 316
compound eye 116
concentration, physical 91
concentrations 41, 68, 83, 217, 233, 234
near-surface 30
concerns of fishers 257
condition index 132, 209, 210, 211
conservation measure 311, 313, 314, 343
consensus 343
consumption of krill
by birds 196
by fish 183±4, 199
by predators 222, 317, 323, 339
by seals 197
by whales 198
continental shelf 63, 74, 75, 204, 256
continental slope 47, 60
conversion factor 223
conveyer belt analogue 68, 72
Copepoda 10, 27, 43, 93, 183, 186, 188, 195
copper 273
copper-calibration sphere, 23
corn gluten meal 269
critical period/distance 213, 215, 323±4
cross-sectional area 96
Crozet Island 302
crystalline cone 116
Acoustic Doppler Current Profiler (ADCP) 73, 94
current, geostrophic 69, 71, 75
current surface 73
Cumberland Bay 68, 74, 217
cycle, diurnal 89, 92, 93, 94
Cysteine 271
Davis Strait 4
Decapoda 1, 183
Deception Island 5, 63
decision rule 311, 312, 314, 315
deep scattering layer
358
Index
degree of dispersion 89
demersal
environment, Japan 182±185
fish 45, 47, 182, 185
landings 183
demographic change 74
demography 8, 16, 223±4
density 137, 209, 316, 319
absolute 15
biomass 94
numerical 11, 15, 16, 42, 47, 54±8, 60, 66, 70, 73,
75, 239
Department of Fish and Oceans (Canada) 341
dependent species 77, 202, 208, 213, 220, 223, 256,
309, 341, 342, 343
status 224
depleted populations 309
depth, daytime 47, 54
detection using seabirds 247
deterioration 255, 288
detritus 128
development 105, 172, 173
developmental descent/ascent 172
developmental stages 42
digestive enzymes 263
dinoflagellate 188
direction of tow 239
discount factor 314
`Discovery investigations' 8, 12, 34±5, 63, 118
`Discovery' measurement' 35
discrete targets 22
discrete sampling (patches) 82
distribution 137, 190, 328
depth 59, 96
horizontal 59, 66
mesopelagic 3
mesoscale 68
micro-scale 82
seasonal 306
vertical 59
diurnal
variation 241
migration model 93
diver observation 8, 6, 89, 94, 96, 98
Dixon entrance 53
dodecahexaenoic acid (DHA) 275, 276, 295
dogfish, spiny (Squalus acanthias) 190
downstream effects 74
Drake Passage 69
drifter buoys 217, 304, 317
early larvae 42
Eastern bloc 229
echochart/echogram 48, 82, 186, Plate 3
echosounder 10, 13, 14, 48, 238, 239, 244, 247, 256
echotrace 249
eclipse, lunar 94
ecological
considerations 342, 344
importance 307
modellers 341
relationships 309, 341
risk, potential 309
economic
analysis 329, 330
driving forces 329
economics, fishery 277, 294, 342, 344
ecosystem
approach 76, 223±224, 309, 341
assessment 81, 91, 202, 220
changes in 309
considerations 330
dynamics 202
effect on 341
interactions 38, 97, 341
monitoring 199
monitoring and management 219±223
eddying effects 74
eel (Anguilla spp) 268
egg 42, 105, 136
abundance/concentration 48, 157
counts 48, 294
germinal zone 162
perivitelline space 172, 173
production 71, 153, 165, 174
released 170, 171
spawned, number 49
volume 172
Ehime Prefecture 296
eicosapentaenoic acid (EPA) 275, 276, 295
ejaculatory duct 36, 37, 38, 158
Elephant Island 68, 70, 72, 73, 84, 90, 91, 139, 156,
218, 304, 322
El NinÄo Southern Oscillation (ENSO) 53 136, 133,
190, 291
embryo size 171, 174
Emerald Basin 60
emigration 107
energy
balance 184
budget 117
costs, reproduction 152, 153
flow 186±188
reserves 171
environment 54, 71, 202, 306, 318, 321
information 258
physical 93, 138
environmental
change 224
conditions 125, 156
factors 234
parameters 135
variability 161
endogenous rhythm 94
endopod 35
English Channel 4
ensonified volume 20, 25
enzymes 276, 277
equator 3, 4
euphausiids 1, 27, 28, 43, 54, 183, 192, 268
identification 1±2
Exclusive Economic Zone (EEZ) 228, 326
Index
excretion, ammonium 126
exopodite 2
exoskeleton 33, 106, 123, 264, 302
eye 2, 5, 34,
-ball 116
stalk 33
FAO 64, 74
Far East Polar Vortex index 291
fat body 161, 165
fatty acid 263, 266, 269, 270, 273, 275, 276
profile 272
synthesis 123
fecundity 167±73
feeding 80, 92, 98
cannibalism 136
condition 70, 198, 234, 244, 258, 221
grounds 204
intensity 204
pressure 133
requirements 98
status 210
female 35, 38, 126, 128, 129±131, 132, 152, 159
FIBEX 65, 66, 76, 304, 311, 314, 315, 317
field-based experiments 105
Fine Resolution Antarctic Model (FRAM) 68, 76,
217
First International BIOMASS Experiment: see
FIBEX
fish 93, 194, 204, 206, 208, 210, 229, 232, 249,
256±7, 323, 325±6 341, 343
meal 269, 278
schools 80, 96
size 183
stocks 75
stomach contents 247
fishing (see also nets and vessels)
conditions 251, 290±1
depth 251
duration 325
effects on dependent species 320±1
effects of localised 314
exploratory 300, 304
fleet organisation 232±33, 246
ground 228, 233±8, 246±7, 256±7, 286, 288, 290
methods 257
pattern 306, 322
pelagic 228
period 294, 295
ports 246
pressure 224
pulse 325
season 42, 45, 48, 247, 251, 255, 257, 285, 286,
290, 313
strategy 228, 232, 242, 253
fisheries, alternatives 257
expansion, potential for 341
management 173±175, 224, 309, 341
managers 343
operations 138
`unconventional', 300
359
fishery 61, 67, 202, 211, 215, 223±4, 228
Antarctic 263, 265
forage 278, 341, 343
history 228
status 286±288, 339
fixation
alcohol 263
formalin, 128
fjord 105, 107, 128
flatfish 183, 190
fleet organisation 228, 255
flow fields 68, 75
flowmeter 13
fluctuation, natural 342
fluoride 264±6, 269, 302
flux 75, 76, 105, 216±218, 311, 317, 318, 323, 328
food 93
availability/production 122, 125, 126, 153, 156
deprivation 106, 125
limitation 136
scarcity 116, 127
supply 123, 80, 93
foodweb 74, 157
Antarctic 194±9
Canada 190±2
Japan 182±188
formalin fixation 128
France 230
freezing capacity 255, 339
freshwater discharge 57
frozen krill, see products
Fukushima Prefecture 46, 50, 182, 184, 244±6, 248,
285, 290, 293
fulmar, Antarctic 92, 196
functional relationship 207, 315, 322
Fundy, Bay of 56, 165, 339
Fundy Channel 60
furcilia 41, 44
gadoid fish 183
gametogenesis 162±4
Gaspe Island 58, 59
Generalised Yield Model (GYM) (see also Krill
Yield Model) 203, 312
genetic diversity 342
geographical information system 29, 30
Georges Bank 191
Georgia, Strait of 55, 190, 252, 253, 256, 339
geostatistics 86
Germany, East (GDR) 229, 230
germinal zone 162
Gibralter 4
gills 2, 37, 38
global warming 140
GLOBEC,
glucosamine 269
Glycerophospholipids 276
glycine 269
glycoproteinic yolk 161, 163
gonad,
maturation 210
360
Index
production 123
resting 165
size 209
Grand Banks 57, 192
greatest net annual increment 309
Greenland 4, 110, 112, 114, 119, 192
growth 74, 80, 105, 109, 116, 138, 267, 311
efficiency 117
increment 122, 123
K (VBGF), 118, 135
maximum 118, 122
multiple morphometrics 115±16
period, annual 118, 122
rate 105, 108, 117±23, 125±6, 133, 152
daily 117, 118, 122
instantaneous (IGR) 117, 119±121, 123,
125±126, 127
potential 122
performance 123
promoter 269
seasonal 312
shrinking 106±140, 238
von Bertalanffy function (VBGF) 118, 123, 124,
135
zero 118
gull, black tailed (Larus crassirostris) 182, 187, 188
Gullsmarsfjorden 93
gyre 40, 75
haddock (Melanogrammus aeglefinus) 191, 342
hake 54, 213, 341, 342, 343, 344
longfin (Urophycis chesteri) 192
Pacific (Merluccius productus) 190
silver (Merluccius bilinearis) 191
hakeling, threadfin (Laemonema longipes) 182, 183
Halifax workshop 341, 342
harvesting 208, 228±258, 310, 343, 344
at two trophic levels 343±4
Hatteras Cape 4
haul by haul data 67, 74, 257, 258, 322
haul duration 239
Heard Island 302
Hecate Strait 53, 190
helminth parasite 264
hepatopancreas 38, 39, Plate 1
herbivores 122
herring (Clupea harengus) 191, 342
Pacific (Clupea pallasi) 184, 187, 190
histidine 271
Hokkaido 46, 295, 296
Honshu 40
Howe Sound 252, 253
hydrolytic enzymes 263
Ibaraki Prefecture 44, 45, 46, 244, 245, 246, 248,
251, 252, 285, 288, 290, 291, 293
ice 204
algae 126, 138
associated biota 70
-bergs 233
concentration 301, 319
conditions 140
coverage 140
cover 302, 307
edge 234, 235, 306
bloom 156
zone 158
free area 306
days 308
retreat/return 308
-zone, marginal 127
icefish, mackerel (Champsocephalus gunnari)
195±6, 199, 204, 208, 211±3, 223, 326±7
Iceland 4, 112, 119, 152
immature 44
immigration 107
impact, monitoring of 342
incubation shift 211
index
oceanographic
Far East Polar Vortex index 291
fishing to predation index 223
Indian Ocean 1, 14, 63, 65, 66, 67, 72, 109±12, 117,
121, 124, 125, 132, 134, 139, 152, 229
Indices
integration period 212
of krill abundance 316
of predator performance 320
Industrial information from fishery 329
in situ observation (patches) 82, 86
interactions 202
large scale 203
small scale 203±206
spatial and temporal scale 199
International Commission for the Conservation of
Atlantic Tunas (ICCAT) 329
International Commission for the Southeast
Atlantic Fisheries (ICSEAF) 329
Inubo-zaki 285
Irish Sea 4
iron 273
irreversible decline, risk of 341
Ishikari±Shiribeshi waters 295
Ishinomaki 46
fish market 183, 184
isoleucine 271
isolume 93
isotherm 3
Italy 66
iteroparous 154
Iwate Prefecture 46, 244±6, 248, 252, 285, 288, 291,
293
Japan 1, 40, 45, 111±2, 114, 118, 121±2, 128, 136,
204, 228, 229, 230, 233, 236, 241, 248, 250,
256, 267, 270, 277, 284, 303±4, 316, 319,
322
Japan Sea 3, 40, 41, 284, 285
Jervis Inlet 55, 56, 253, 256, 339, 340
Joban area, Joban Prefecture/Waters 40, 41, 42,
48, 185, 187±8, 244±7, 251, 285±6, 290, 297
Jordan Basin 60
Index
jurisdiction, national 343
juvenile 36, 37, 42, 43, 44, 69, 71, 117, 118, 125, 132,
133, 138, 319
361
Kamaishi 45, 46
Kamchatka 3, 111, 114
Kara Sea 4
Kashima-nada Sea 42, 184, 289, 290
Kattegat 4, 96, 112, 117, 122, 128, 129±131
Kemp Land 234, 235, 238, 239, 243
Kerguelen Island 302
ketones 263
King George Island 68, 72, 218, 322
Kinkasan Island 45, 46, 183, 291
Knight Inlet 253
Korea 229±30, 303±4, 307
krill
escapement 310, 312, 315, 316, 323
season, poor 207
surplus 228
uses 265±278
Krill
Yield Model (KYM) 203, 223, 224, 312
gamma 203, 311, 312, 314
lambda 203
Kuroshio
current 3, 136, 289, 290
Extension/Front 40±3, 291
leucine 271
level of harvest 311
license 246, 255, 288, 292, 293, 339, 343
life
cycle 122
multi-year 108
expectancy 114
history model 127
-span 107, 108, 109±27, 133, 135, 150, 151, 152
light 93
Ligurian Sea 165, 166, 167, 170
lipid 266, 270, 275
biochemistry 157
content 126, 269, 270
neutral 157
synthesis 118
total 152
lipidic yolk 161
lipofuscin 115
Livingston Island 139, 322
longevity 105, 110, 133, 135
Longhurst±Hardy Plankton recorder (LHPR) 10,
12, 86
long lived species 108
Low Island 322
LTER±70
luminous organ 2
lysine 271
laboratory studies 105, 123, 158
Labrador 4
current 157
lactation 267
lamps, fish-gathering 296
lantern fish 185
larvae 118, 127, 133, 136, 138
larval abundance 53
larval mortality 135, 174
Lasarev Sea 155
Latvia 229, 230
law of the sea 228, 326
Laurentian Channel 58, 59, 60, 91
layer 28, 59, 86, 90
La Have Basin 60
length
asymptotic 118
at age 109, 112, 114
based ageing 106
density 14, 108
frequency 13, 15
analysis 106, 107, 116, 126
distribution 108, 109, 117, 127, 137, 218, 219
mixtures 108±9
polymodal 15, 106, 107, 108
unimodal 15
maximum, total 152
mean 109, 127
measurement 33
modal 117
total 132
-weight relationship 128±132
mackerel 191, 296
jack 80
Japanese chub, (Scomber japonicus) 182
Magdalen Shallows 57, 58
magnesium, 273
Maine Gulf of 4,
maintenance ration 106
Malaspina Channel/Strait 55, 253, 256, 339, 340
male 126, 128, 129±131, 132, 152, 159, 295
mammals, latitudinal distribution 197
management 50, 60, 202, 208, 223, 224, 328
action 223
advice 74
areas 74, 231, 302, 310, 318
approaches 309±10
cost of acquiring knowledge 342
criteria 343
feedback 325, 328
fisheries 173±175, 224, 309, 341
measures 325
national jurisdiction 343
objectives 309
paradigm 311, 320
plan 341
predictive 325
principles 309
procedures 314
performance of 310
reactive 325
regime 307, 342
regions 314
Southern Ocean 300±330
362
Index
Japan 284±297
regulations 293, 294
risk 222
units 74, 77
Marion Island 302
Maritimes Region 60
market
limitations 258
price/value of krill 294, 295, 297, 344
mark-recapture 105
mating 158±160
behaviour 158, 159
chase 158, 159
timing 158
maturation 118, 135, 319
ova 154
testis 161
mature 82, 113, 295
maturity 11, 13, 35, 37, 57, 70, 104±6, 110, 114, 115,
126, 128, 152, 234
keys 35±8, 161±166
size at 81
maximum sustainable yield 203
McDonald Island 302
median krill escapement 312
Mediterranean Sea 5, 110, 114, 123, 165, 228
measurement
carapace 129±132
`Discovery', 35
reference 33, 35
standard 33, 35
metabolic rate 126
methionine 271
mid-water zone 44
migration 70, 71
diel 24
feeding 213
minerals 273
mixed layer depth 188
temperature 44
Miyagi Prefecture 46, 184, 244, 245, 246, 248, 251,
285±88, 290, 291, 293
M/k ratio 135
molecular biology 75
monitoring effects of harvesting 320±326
monitoring programme 343
morphology 33
morphometrics 104, 132
mortality 74, 80, 106, 127, 133, 135, 218
fishing 309, 320
F0.1, 309
larval 174
natural 108, 203, 220, 222, 223, 312
overwinter 107, 133±135, 311, 312
rate 152, 317
moult 81, 82, 106, 125
cycle 117, 122±3, 125, 132, 168, 170
process 105, 123
multispecies applications 309
murre, common (Uria aalge) 186
thick-billed (Uria lomvia) 186, 190
murrelet, ancient (Synthliboramphus antiquum)
186
Mysidacea 1, 10
Nagasaki Prefecture 285
Nakaminato 45, 46
near-bottom water 47
Nelson Island 322
neritic 63
net 60, 66, 244, 339
avoidance 12, 13, 15, 28, 89
bongo 12
bow mounted trawl 249, 250, 251, 285, 286,
294
catches, plankton 89
depth 13, 239
fine-mesh 104
gill 244
hauls 63, 69, 72, 73, 218, 219, 319
larval 42
MOCNESS 11, 12, 43
monitoring 22
NORPAC 41, 48
Ocean Research Institute 41
plankton 10
purse seine 296
Rectangular Midwater Trawl (RMT) 10, 12
sampling 8±16, 26, 54, 82, 104
scientific 10
Seine 246, 249, 250, 255, 288, 296
selection 11, 104, 108, 133, 138
selectivity 10, 11, 12, 16, 218±219
size 12
-sonde/sounder 14, 240
spoon 295
status 13
towing 12
trawl 239, Plate 5
beam 41, 45, 47
-tow, random 14
New Zealand 107, 234
night-time depth 47
night-time sampling 13
NOAA 247
Norpac net 41, 48
North Pacific Krill Resources Research Group
42
North Sea 109, 110, 119, 123
Norway 110, 111, 112, 113, 114, 118±20, 123, 124,
128, 129±131
Norwegian Sea 4, 5
Nova Scotia 1, 57, 109, 110, 112, 129±131, 191, 192,
339
nuclease 263
oasis effect 156
oblique tow 14
Observer Programme (CCAMLR) 34, 37
observations in situ of patches 82, 86
ocean currents 105, 217
oceanic circulation 69
Index
oceanic deep water 63, 68
oceanographic
conditions 285, 286, 288±90, 292
frontal area 42
octopus 183
Okhotsk Sea 3, 40, 185, 186, 296
omnivores 122, 136, 157
Onagawa 44, 291
fish market 285
oocyte 162
mature 164
maturation 163
young 161
oogenesis 161, 163
oogonia 162
open access fisheries 326
open/closed areas 325
optical plankton counter (OPC) 8, 10, 11
Oregon 111, 112, 114, 120, 122, 133, 134, 136,
157
Osake Peninsula
orientation 24, 26, 47
Oshika Peninsula 244, 245, 285
Otsu 291
fishing ground 252
ova maturation 154
ovarian
development 153, 154, 156
maturation 157, 160±1
reorganisation 163
ovary 38, 159
ovary, gravid 126, 128, 132, 158, 160, 174
over-exploitation 326
overlap, krill fishery and predators 211, 213±216,
220, 222±223, 320, 323, 325
Agnew and Phegan 215, 222, 323
fishing to predation 223
Schroeder's 215, 223, 301
potential 213, 323
precautionary 213
realised 215
overwintering strategy 128
oviposition 163
ovulation 161
Oyashio
branch 41, 292
current 40, 42, 136, 244, 246, 247, 251, 252, 256,
285, 288, 289, 290, 291
extension 40
front 40
ozone depletion 140
Pacific Ocean,
Antarctic sector 1, 63, 229
North 3, 4, 40, 45, 53, 63, 111±3, 122, 160, 204,
229, 264, 291, 341
pack ice 66, 197, 235, 236, 237
Palmer Long-Term Ecological Research Program,
see LTER
Panama 229, 230
parasites 264
363
parental
input 172
stock density 136
patch 59, 72, 75, 81, 83
classification scheme 81, 82, 83
physical parameters 81±2
zooplankton 186
patches 22, 24, 25, 28, 29
patchiness 11, 29, 30, 80, 137
particle counter 13
pelagic environment 185±186
penguin 204±6, 211, 213, 322, 325
activity 212
Adelie (Pygoscelis adelie) 195±6, 206, 315, 322
chinstrap (Pygoscelis antarctica) 195±6, 206,
322, 323
fledging weight 207, 211
gentoo (Pygoscelis papua) 195±6, 205, 219, 323
macaroni (Eudyptes chrysolophus) 16, 195±6,
205, 215, 219±20
perch, yellow (Perca flavescens) 268
perivitelline space 172, 173
petasma 36, 37
petrel 204, 206
Antarctic (Thalassoica antarctica) 195, 196,
206
Cape (Daption capensis) 92, 195±96 206
diving (Pelecanoides spp) 195±6, 204
giant (Macronectes spp) 195±6
Leach's storm (Oceanodroma leucorhoa) 192
snow (Pagodroma nivea) 195, 196
storm (Oceanites spp) 195
white-chinned 196
Wilson's 196
phenylalanine 271
phospholipase 263
phosphorus 273
photographic sampling 8,10
photographic estimates of density 89
photophore 2
photoperiod 154, 157
phototaxis 59, 91
phytoplankton 38, 57, 58, 63, 136, 138, 153, 156,
157, 187, Plate 1
pigments 270, 272
carotenoid 266, 268, 269, 270, 272, 275
plankton
plant protein 269
pleopod 2, 35, 36
Poland 229, 230, 266, 303, 304, 307, 319
polar front 235
polar species 153
pollock 213
walleye (Theragra chalcogramma) 47, 48,
182±4, 187, 249, 268
polynya 156
population 75
age structure 106±27
captive 10
dynamics 74, 127, 310
growth rate 126
364
Index
harvested 309
parameters 103
processes 105
reproducing, percentage 167
self-regulating 174
port facilities 67
post larval
life cycle 135
population 133,
potassium 273
potential yield 317
predation 72, 80, 136, 138, 218, 323
impact 183
pressure 54, 223
risk 98
predator 16, 72, 76, 85, 91, 93
avoidance behaviour 88, 91, 98
breeding 72, 204, 207, 211, 320
defence against 92
demands 135
diet 210, 222, 223
diving depth 204
energetic requirement 222, 223
food for 310
foraging
activity 205, 213
behaviour 203
distribution 215
footprint 215, 221, 222
local prey abundance 217
range 205, 218, 310, 322
success 320
trip duration 211
index 207, 320
predator, land-based 218, 314
needs 314
performance 315
population size 222, 223
prey interaction 92, 212
predators 104, 108, 133
Prefecture (Japanese administrative region)
Akita 285
Ehime 296
Fukushima 46, 50, 182, 184, 244, 245, 246, 248,
285, 290, 293
Ibaraki 44, 45, 46, 244, 245, 246, 248, 251, 252,
285, 288, 290, 291, 293
Iwate 46, 244, 245, 246, 248, 252, 285, 288, 291,
293, Plate 3
Joban 40, 41, 42, 48, 185, 187±8, 244±7, 251,
285±6, 290, 297
Miyagi 46, 184, 244, 245, 246, 248, 251, 285, 286,
288, 290, 291, 293
Nagasaki 285
Sanriku 40, 41, 42, 46, 47, 48, 182, 184, 185, 187,
188, 244, 245, 246, 247, 251, 285, 286, 287,
289, 291, 292, 294, 297
Yamaguchi 285
pre-spawning stage 128
previtellogenesis 163, 164
price of krill 294, 295, 297
primary production 59, 136, 157, 194
Prince Edward Islands (Antarctica) 302
prion 195, 196
Procellariiformes 192
processing 229, 238, 246, 255, 257, 304, 329, 339
capacity 253
freezing 264
freezing capacity 255, 339
hydrolysates 265, 268, 271, 272
grading 232
peeling 264, 304
products 232, 256, 266, 344, Plates 6±8
Antarctic Krill Concentrate 266, 267
aquaculture products 267, 268
canned tails 266
claimed attrributes of 267
debriding agent 276
diet supplement 267
feed, animal 265
food additive 266±7
freeze dried 252, 265±6, 268, 270, 274
frozen 252
boiled 265
for human consumption 257, 266, 278, 295
fresh 265, 266, 268
quality 232, 239±40, 242, 245, 257, 258, 329
market value 277
meal, 269
anchovy 271
fish 269, 278
krill 264, 267, 271, 344
nutritional value 268
paste 264±6
peeled 232
pet food industry 255
production of 344
protein concentrate 264
shrimp feed 273
storage 263
tail meat 277
uses 265±278
production 68, 216, 222
cycles 155
season 122
productivity 122
proline 269
protease 263, 276
protein 128, 263, 266, 300
content 270
protozoa 126
Prydz Bay 90, 91, 195, 196, 302
pseudopopulation 160
Pte des Monts 58, 59
Puget Sound 153, 157, 167
quality of product 233
Queen Charlotte Islands 53
quota 255
rational use 309
ration, daily 184, 190
Index
rattail fish 183
razorbill (Alca torda) 192
rearing 105
recruitment 14, 53, 71, 74, 108, 133, 135±140,
173±4, 221±2, 311, 318±9, 342
recruits 114, 116, 136
redfish (Sebastes spp) 192, 342
red krill 240, 245
reference points 309
Regional Advisory Process (RAP) 339, 341
regional warming 244
regulations, fishery 293, 294
related populations 309
remotely operated vehicle (ROV) 8, 10
reproduction 98, 150±175
continuous 153
cycle 92, 150±172, 210
energy costs 152, 153
fecundity 167±73
limited 153
secondary sex characters 161
spawning 133, 136, 158±160, 319
activity 42, 58
age at 113, 115, 118
biomass 312, 319
condition 209
duration 153±157
episodes 49, 168, 169, 170
event 107, 128, 132, 136
first season 112
frequency 167
intermittent 153
period 12
population 174
pulses 107
repeat 114
sea-ice effect on 156
second time 113
stock 137, 314, 319
survey 48
timing 153±157
zones 160
reproductive
activity 295
development 172±173
patterns 151, 153
season 69, 107, 132, 138, 151, 154±5, 204±5
success 137, 216
residence time 75, 105
resident population 105
resource partitioning 182
retention time 317
review (conservation measures) 313
rhythm
endogenous 94
risk 341±2
rockcod,
grey (Notothenia squamifrons) 327
marbled (Notothenia rossii) 326, 327
Patagonian (Patagonotothen guntheri) 327
rockfish, spinycheek (Sebastolobus macrochir) 183
365
Ross Sea 63, 66, 155, 198
rostrum 2, 33, 34
Russia/Russian Federation 230, 270, 301, 303, 304,
329
Saanich Inlet 252
salinity 54, 59
salmon 244, 253, 269±70, 278, 341±2
Atlantic (Salmo salar) 268, 272
chinook (Onchorynchus tschawytscha) 190
chum (Onchorynchus keta) 190
coho (Onchorynchus kisutch) 190, 272, 295
Pacific (Onchorynchus spp) 268, 271
sock-eyed (Onchorynchus nerka) 186
Salmonids 272
salps 138, 238
sampling
bias 11, 15
day/night-time 13
density 105
direct 9
interval 104, 105
representative 9, 11, 12, 104
sand lance (Ammodytes personatus) 182, 187, 188,
192, 244, 249, 285
Sanriku Prefecture/Waters 40, 41, 42, 46, 47, 48,
182, 184, 185, 187, 188, 244, 245, 246, 247,
251, 285, 286, 287, 289, 291, 292, 294, 297
Sanuenay River 58
sardine 244, 296
Japanese (Sardinops melanostictus) 183, 185,
187
scale,
temporal 16
spatial and temporal 68, 203, 205, 322
scarcity, of krill 210
scat samples 218
SCAR,
SC-CAMLR 301±330
schooler, facultative 92
schooling 104
parallel orientation 81, 85, 88
schools 22, 25, 28, 47, 83
Schroeder's index 215, 223, 301
SCOR,
Scotian shelf 56, 57, 191, 339
Scotia Arc 318
Scotia Sea 5, 13, 63, 68, 111, 134, 155, 217, 242
Scotland 4, 110, 112, 114, 119, 123, 134
scuba observations 88
sculpin 192
seabass, Australian 268
seabird populations 320
sea-bream (Sparus auratus) 244, 268, 269, 285
Red (Pagrus major) 268, 272, 295, 297
sea-ice 65, 70, 71, 127, 138, 318, 319, 320±1
cover 198, 312
duration of winter 138
dynamics 156, 312
zone 155
seal 206
366
Index
Antarctic fur (Arctocephalus gazella) 16, 91,
194, 196, 197, 204, 206, 215, 218, 219, 223,
315, 316
crabeater (Lobodon carcinophagus) 194, 196,
197, 204, 206, 315
diving range 196
elephant (Mirounga leonina) 196
leopard (Hydrurga leptonyx) 194, 196, 197
populations 320
size 197
Ross (Ommatophoca rossi) 196
Weddell (Leptonychotes weddelli) 196
sea-mount 91
seasonal
change 41, 47, 55, 72
cycle 107, 116, 122
sea-surface 97
temperature 140
segregation, spatial 159
Sendai Bay 46
sequential hauls 68
setae 33, 34
sex 104, 126
ratio 86, 152, 158, 160
sexual
appendages 36
characteristics 106
Shakotan Peninsula 295, 296
shearwater
pale footed (Puffinus carneipes) 185
slender-billed (Puffinus tenuirostris) 185, 186,
187
sooty (Puffinus griseus) 185, 187
streaked (Calonectris leucomelas) 185
shelf 60, 63, 67
break 54, 56, 72, 75, 91, 204, 234, 306
waters 91
shellfish 256
shelter for fishing vessels 67
Shikoku 296
Shimane 285
ship, mother 233
ship operation 232
shrimp 271
shrinkage 106, 116, 125±7
similarity matrix 14
simulation, stochastic 311
sinking rate 172, 173
size 11, 198, 234, 238, 240
at age 109
at first spawning 150±152
classes 322
maximum 114, 117, 122
at maturity 151
selection 108
composition 133
frequency 34
Skagerrak 4, 111, 112, 114, 117, 122
social interactions 81±2, 92
sodium 273
solubility, of krill protein 263
somatic growth 123
sound scatter 24, 26, 29
South Africa 230
South America 234, 235
South Georgia 13, 16, 68±9, 74, 85, 90±1, 111,
160, 195±7, 205, 208, 215, 217, 223, 229,
234±5, 237±8, 240±3, 302, 304, 306, 313,
317±8, 320
South Orkney Islands 68, 69, 234±5, 237±8, 240±3,
302, 304, 313, 318, 320
South Shetland Islands 47, 63, 67±70, 72, 138,
217±8, 222, 234±5, 237±243, 302, 304, 313,
318, 322±5
South Sandwich Islands 313
Southern Ocean 1, 2, 4, 14, 63±4, 66, 74, 126, 140,
195±6, 204, 220, 229, 342
Soviet Union 301, 303, 316, 319
soybean protein 269
Spain 119, 229, 230, 266
spatial
coverage 16
distribution 67
pattern 29
resolution 26
scale 81, 82, 322
separation 69
spawning (see reproduction)
Specialty Marine Products, Vancouver 267
spectral analysis 28, 30
spermatophore 35, 36, 37, 38, 152, 158, 295
sperm mass 158
Spitzbergen 160
split year 229, 301, 302, 303, 319
spoilage 263
sport fishing 274, 285, 295
spring bloom 153
squid 186, 194, 195, 199, 206, 232, 234, 244, 257,
323
St Lawrence Gulf of 4, 56, 57, 58±60, 91, 129, 191,
339
stable recruitment 309
stakeholder 343, 344
standard measurement 33, 35
standing stock (biomass) 65, 66, 68, 72, 294, 339
starvation 106, 127
statisticalareas 231, 302
subdivision of 74
Statistical Bulletin 301
status of krill stocks 310
steroids 269
stock 310
biomass 314
density 137
-recruit relation 135, 136, 173
size 135, 140, 311, 318
stomach
contents 47, 183, 209, 218, Plate 4
fullness 209
stone crab 195
stress 105
strobe lighting 11
Index
sub-adult 37, 38, 69, 71
subAntarctic zone 5
subarctic 110, 128, 152, 153, 157
boundary 41
front 41
subarea scale 68
submersible vehicle 45, 46, 185
sub-optimal diet 106
subpolar species 153
sulfur compounds 263, 273
summer growth rate 117
super-swarm 72, 83, 86, 90, 91
surface layer 98
survey 67, 104, 311, 312, 318, 341
abundance estimation 252
design 66, 91
plankton 294
research 68, 211, 223
small-scale 105
spawning 48
synoptic 65, 315
track 14
survival 49, 136, 138
Suruga Bay 40
swarm 1, 5, 10, 11, 14, 15, 28, 42, 47, 63, 68, 73,
80±98, 188, 239, 240, 247
composition 85
cross-sectional area 94
depth 86, 97
density 88, 89
detection 238, 247±51, 256
discrete 90
group size 88
mid-water 44
nearest neighbour distance within 88, 89
physical structure 86
shape 86, 96
similarity between 82, 84
subsurface 286
surface 45±6, 92, 96, 159, 174, 239, 247, 257, 286,
294
thickness 86
swarming 174, 216, 290
daytime surface 98
season 295
symbionts of krill 263
Taiwanese waters 297
Tasmania 1, 5, 80, 185, 276
target fishing 14,
target species (of fishery) 76, 202
target strength, see acoustic target strength
Tartary, Gulf of 40
telson 2, 33, 34
temperature 59
air 140
limit 40
compensaton 122
minimum, annual 286
optimal 40
for aggregation 252
367
optimum 122, 290
sea surface 42, 44, 233, 246±7, 248, 287, 288, 289,
291, 294
water 45, 47, 137, 291
temperate water species 107, 126, 153, 154, 155
temporal
change 16
pattern 29
scale 16, 81, 82, 221
succession 70
terminal setae 33
testes 161
thelycum 36, 37, 158, 162, 295
therapeutic procedures 276
thorax 2, 36
threonine 271
Tilapia 271
tilt angle 27
Tokyo 46
toothfish,
Patagonian (Dissostichus eleginoides) 195, 314,
327
Antarctic (Dissostichus mawsoni) 195
topographic effects 60
Total Allowable Catch (TAC), see catch, total
allowable
transducer 10, 11, 21, 22, 23, 24, 98, 239
transects 65, 66
transparent krill 232
transition zone 42
trawling
information 258
depth 239, 240, 241, 242
time 239
triacylglycerol 152
trophic analysis 184
tropical water euphausiids 93, 126
trout, rainbow 271, 272, 295
Tsugura Warm Current 41, 42, 291
Tsushima current 41
Tsushimi Islands 285
tyrosine 271
Ukraine 230, 301, 303, 329
uncertainty 310, 312, 328
allowance for 312
under-ice environment 97, 98
unimodal size distribution 106, 107
United Kingdom 229, 230, 303±4
unspecified marine crustacea 301
upwelling 59, 91, 122, 136, 157, 251
uropod 2, 33, 34, 125
usage of krill 295
USSR 228, 229, 230, 233, 256
Uwajima Bay 265, 296±7
valine 271
Vancouver Island 53, 54, 55, 190, 252, 253, 338
variability,
environmental 161
heterogeneity 14, 80
368
Index
heterogeneous distribution 11
inter-annual variation 13, 53, 140, 249, 287
small-scale 14
variation
latitudinal 69
spatio-temporal variation 77
inter-annual in year class strength 217
vasa deferentia 35
vernacular names of krill
Ami-ebi 296
Tsunonashi-okiami, 284
vertical
distribution 205
migration 80±98, 216
diel 24, 47, 54
diurnal 9, 196
vessel, cargo 233
vessel, fishing 239, 257, 288, 292, 294, 325
vessel freezer 255
vessel, scouting 233
vessel, trawler 233, 239
vessels, number of 255
video 8, 13, 45
viral infection 264
visual
cues 9
estimates of density 89,
observations 88, 215
sampling 8, 10,
sighting, of swarms 247
vitellogenesis 160, 163, 164
von Bertalanffy growth function (VBGF) 118,
123, 124, 135
warm
core ring 40, 41, 43, 44, 291, 292
water 136
species 44, 107
tongue 43
Washington 111, 119, 120, 136, 153
water
circulation 72, 77, 318
column 11, 127
mass 291
turbidity 57
temperature 45, 47, 137
Weddell Sea 109, 112, 124, 155, 156, 302, 313
weight 132,
dry 123, 128, 129±131, 132
wet 49, 123, 128, 129±131, 262
western sub-Arctic gyre 41
whale 60, 91, 194, 204, 206, 228, 323
blue (Balaenoptera borealis) 190, 197, 198
body fat 198
bowhead (Balaena mysticetus) 191
distribution 197
feeding grounds 60
fin (Balaenoptera physalus) 190, 197, 198, Plate
4
herding behaviour 91
humpback (Megaptera novaengliae) 190, 197
minke (Balaenoptera acutorostrata) 185, 187,
188, 197, 198
sei (Balaenoptera borealis) 197
Southern right 91
whitefish 268
White Sea 4
Wilkes Land 234, 235, 236, 238, 239, 243
wind 291
wind speed 239
winter shrinkage 107
winter feeding 123
Working Group (SC-CAMLR) 207, 220, 301
Workshop, Vancouver 339
Yagishiri Island 295, 296
Yamaguchi Prefecture 285
Yellow Sea 3
yellowtail (Seriola quinqueradiata) 268, 272, 295
yield 311, 312
long-term 311
krill model (KYM) 203, 223, 224, 311±4, 321
per recruit 309
yolk accumuluation 165
younger stages 104
zero growth 118
zinc 273
zooplankton 26, 27, 28, 29, 47, 54, 63, 126, 186, 190,
339
feeders 183
patches 186
Taxonomic Index
Euphausiids
N. simplex 107
Bentheuphausia amblyops 4
Tessarabrachion oculatum 3
Euphausia
E. crystallorophias 5, 63, 64, 105, 150±2, 155, 157,
164, 168±9, 173, 195
E. eximia 107
E. frigida 5, 63, 64, 150, 155, 195
E. gibboides 3, 43±4
E. hanseni 168
E. hemigibba 3, 43
E. longirostris 5
E. lucens 88, 89, 105, 150±3, 155, 160, 168±9, 172
E. nana 3, 150±1, 173, 228, 265, 284, 296, 297
E. pacifica 3,28, 40±5, 47±50, 53±5, 88, 105±8,
111±2, 114±5, 118±122, 124±5, 128, 130±131,
133±6, 150±4, 157±8, 164, 167±9, 173,
182±190, 228, 244- 255, 257±8, 265, 266, 267,
269±72, 276, 277, 284±86, 289, 290±97
E. paragibba 3
E. similis 3, 42, 43
E. superba 5, 11, 14±6, 28, 33, 47, 63±6, 74, 80, 82,
84±6, 88±94, 97, 105±9, 111±12, 115, 117±9,
121±8, 130±5, 137, 139±140, 150±2, 155±161,
163±170, 172±4, 195, 228, 242, 244, 257±8,
262±64, 272, 275±77, 285, 295, 300, 313, 327,
Plate 1
E. triacantha 5, 63, 64, 107, 122, 150, 155, 195
E. vallentini 5, 63, 64, 88
Thysanoessa
spp. 114±5, 122, 133±4, 152±3, 276
T. inermis 3 ,4, 56±8, 106, 108±110, 112, 114,
117±120, 122±4, 127±9, 133±4, 150±2, 154,
157±161, 165, 168, 228, 265, 284, 295, 296
T. longicaudata 56±7, 107, 124, 153, 157, 173
T. longipes 3, 53, 54, 190
T. macrura 5, 63, 115, 150, 155±7, 160
T. oculatum 3
T. raschii 3, 4, 47, 56±9, 91, 93, 108±110, 112, 114,
117±120, 122, 128±130, 133±4, 150±2, 154,
157±9, 172±3, 185±6, 191±2, 264, 265
T. spinifera 3, 53±5, 159, 190,
T. vicina 5, 63
Thysanopoda acutifrons 3
Other taxa
Meganyctiphanes norvegica 4, 5, 11, 47, 56±60, 88,
91±4, 96, 106±7, 110±2, 114, 117, 119, 122,
124, 128±130, 134, 150±4, 157±161, 164±170,
172±3, 191±2, 228, 263±5, 276
Aethia cristaella 186
Alca torda 192
Ammodytes personatus 182, 192, 244, 285
Anguilla spp. 268
Anisakis simplex 264
Arctocephalus gazella 196, 204, 206, 315
Balaena mysticetus 191
Balaenoptera acutorostrata 185, 190, 197
B. borealis 197
B. musculus 190, 197
B. physalus 190, 197
Nematobrachion flexipes 3, 43
Boreogadus saida 190±191
Nematoscelis
N. difficilis 3, 105
N. gracilis 3, 43
N. microps 4, 42
Calanus 339
Calonectris leucomelas 185
Cerorhinca monocerata 182, 185
Champsocephalus gunnari 195, 199, 204±205, 208,
326±7
Chlorophthalmus albatrossis (borealis), 182
Clupea harengus 191, 342
C. pallasi 190, 342
Nyctiphanes
N. australis 5, 80, 88±9, 106±7, 150±3, 155, 159,
168±9, 185, 267, 269, 276
N. capensis 168
N. couchii 4, 107, 150, 153±4, 168
Daption capensis 195, 206
369
370
Taxonomic Index
Dicentrarchus labrax 268
Diomedea chrysostoma 195
D. melanophris 195, 204, 206
Diomedea melanophrys 315
Dissostichus spp 314
D. eleginoides 195, 327
D. mawsoni 195
Eudyptes chrysolophus 195
Gadus macrocephalus 47, 182
G. morhua 191, 342
Hydrurga leptonyx 196
Laemonema longipes 182
Larus crassirostris 182
Lates calcarifer 268
Leptonychotes weddelli 196
Lobodon carcinophagus 196, 204, 206, 315
Macronectes spp 195
Mallotus villosus 191, 342
Martialia, hyadesi 195
Megaptera novaengliae 190, 197
Melanogrammus aeglefinus 191, 342
Merluccius bilinearis 191, 342
M. productus 190, 342
Micropterus salmoides 268, 270
Mirounga leonina 196
Notothenia rossii 326±7
N. squamifrons 327
Oceanites spp 195
Oceanodroma leucorhoa 192
Ommatophoca rossi 196
Onchorynchus spp 268
O. tschawytscha 190, 342
O. keta 190
O. kisutch 190, 295, 342
Ophiura sarsii 185
Pachyptillia spp 195
Pagodroma nivea 195
Pagrus major 268, 269, 295
Paralomis spp 195
Patagonotothen guntheri 327
Pelicanoides spp. 195
Perca flavescens 268
Phoebretia palpebrata 195
Puffinus gravis 192
P. griseus 185, 192
P. tenuirostris 185
P. carneipes 185
Pygoscelis adelie 195, 206, 315
P., papua 195, 206, 323
P. antarctica 195, 206, 322
Salmo salar 268
S. gairdnerii 295
Sardinops melanostictus 183
Scomber japonicus 182
Sebastes spp., 192, 342
Sebastolobus macrochir 183
Seriola quinqueradiata 268, 295
Sparus auratus 268
Squalus acanthias 190
Stizostedion vitreum 268
Synthliboramphus antiquum 186
Thalassoica antarctica 195, 206
Themisto gaudichaudii 209
T. japonica 183
Theragra chalcogramma 47, 182, 249
Todarodes pacificus 183
Uria lomvia 186, 190
U. aalge 186,
Urophycis chuss 191, 342
U. chesteri 192
List of Abbreviations
ACC
ADCP
ANARE
APFZ
AUV
AVHRR
B0
BAS
BIOMASS
BIONESS
CDW
CEMP
CCAMLR
CM
CPD
CPH
CPUE
CSI
DFO
DHA
EEZ
EPA
ENSO
FAO
FFT
FIBEX
FRAM
GIS
GLOBEC
GRT
GZ
ICCAT
Antarctic Circum-Polar Current
Acoustic Doppler Current Profiler
Australian National Antarctic Research Expedition
Antarctic Polar Frontal Zone
Autonomous Underwater Vehicle
Advanced Very High Resolution Radiometer
Pre-exploitation biomass
British Antarctic Survey
Biological Investigations of Marine Antarctic Systems and Stocks
Bedford Institute of Oceanography Net and Environmental
Sampling System
Circum-Polar Deep Water
CCAMLR Ecosystem Monitoring Programme
Commission for the Conservation of Antarctic Marine Living
Resources (also sometimes used for Convention for the
Conservation of Antarctic Marine Living Resources)
Conservation Measure (CCAMLR)
Critical Period Distance (CCAMLR)
Catch Per Hour fishing
Catch Per Unit of Effort
Combined Standardised Index
Department of Fisheries and Oceans (Canada)
Dodecahexaenoic acid
Exclusive Economic Zone
Eicosapentaenoic acid
El NinÄo Southern Oscillation
Food and Agriculture Organisation of the United Nations
Fast Fourier Transform
First International BIOMASS Experiment
Fine Resolution Antarctic Model
Geographical Information System
Global Ocean Ecosystems Dynamics (an international programme
having individual national programmes such as US-GLOBEC)
Gross Registered Tonnage (shipping)
Germinal Zone
International Commission for the Conservation of Atlantic Tunas
371
372
List of Abbreviations
ICSEAF
IGR
k
KYM
LTER
LLHPR
M
MIX
MLMIX
MOCNESS
MSY
NOAA
NND
OPC
PVS
R
RAP
RMT
ROV
SCAR
SC-CAMLR
SCOR
TAC
TS
UNCLOS
UV-B
VBGF
WG-EMM
International Commission for the Southeast Atlantic Fisheries
Instantaneous Growth Rate
Coefficient of catabolism in VGBF
Krill Yield Model (CCAMLR)
Long Term Ecological Research Programme (USA)
Large Longhurst-Hardy Plankton Recorder
Coefficient of natural mortality
Interactive computer programme to analyse a length frequency
distribution and determine modal values
An extension of MIX to provide a length density distribution
Multiple Opening and Closing Net and Environmental Sensing
System
Maximum Sustainable Yield
National Oceanographic and Atmospheric Administration (USA)
Nearest Neighbour Distance
Optical Plankton Counter
Perivitelline space
Recruitment
Regional Advisory Process (Canada)
Rectangular Midwater Trawl
Remotely Operated Vehicles
Scientific Committee for Antarctic Research
Scientific Committee for the Conservation of Antarctic Marine
Living Resources
Scientific Committee for Oceanic Research
Total Allowable Catch
Target Strength
United Nations Commission on the Law Of the Sea
Ultra-Violet light in the frequency range 280±350 nannometres
von Bertalanffy growth function
Working Group on Ecosystem Monitoring and Management
(CCAMLR)
Fish and Aquatic Resources Series
Series Editor: Professor Tony. J. Pitcher
Director, Fisheries Centre, University of British Columbia,
Canada
The Blackwell Science Fish and Aquatic Resources Series is an initiative aimed at
providing key books in this fast-moving field, published to a high international
standard.
The Series includes books that review major themes and issues in the science of
fishes and the interdisciplinary study of their exploitation in human fisheries.
Volumes in the Series combine a broad geographical scope with in-depth focus on
concepts, research frontiers and analytical frameworks. These books will be of
interest to research workers in the biology, zoology, ichthyology, ecology, physiology of fish and the economics, anthropology, sociology and all aspects of fisheries. They will also appeal to non-specialists such as those with a commercial or
industrial stake in fisheries.
It is the aim of the editorial team that books in the Blackwell Science Fish and
Aquatic Resources Series should adhere to the highest academic standards through
being fully peer reviewed and edited by specialists in the field. The Series books are
produced by Blackwell Science in a prestigious and distinctive format. The Series
Editor, Professor Tony J. Pitcher is an experienced international author, and
founding editor of the leading journal in the field of fish and fisheries.
The Series Editor and Publisher at Blackwell Science, Nigel Balmforth, will be
pleased to discuss suggestions, advise on scope, and provide evaluations of proposals
for books intended for the Series. Please see contact details listed below.
Titles currently included in the Series
1. Effects of Fishing on Marine Ecosystems and Communities (S. Hall) 1999
2. Salmonid Fishes (Edited by Y. Altukhov et al.) 2000
3. Percid Fishes (J. Craig) 2000
4. Fisheries Oceanography (Edited by P. Harrison & T. Parsons) 2000
5. Sustainable Fishery Systems (A. Charles) 2000
6. Krill (Edited by I. Everson) 2000
7. Tropical Estuarine Fishes (S.J.M. Blaber) 2000
For further information concerning books in the series, please contact:
Nigel Balmforth, Professional Division, Blackwell Science, Osney Mead,
Oxford OX2 0EL, UK
Tel: +44 (0) 1865 206206; Fax +44 (0) 1865 721205
e-mail: nigel.balmforth@blacksci.co.uk